Consolidated amendments to the manual on VDL · Web viewDRAFT amendments to the manual on VHF...

Appendix GACP-WGM/10 ReportAppendix G

DRAFT amendments to the manual on VHF Digital Link Mode 3 (Doc. 9805, first edition)

Part I – Implementation aspectsChapter 3 – Formats

(11 pages)document.doc

- 3 –

3.3.5The meanings of the timing states of TS1, TS2 and TS3 were discussed briefly in 1.4.5 and will be discussed in detail in Chapter 8. Here it is sufficient to note that the normal state of VDL Mode 3 is TS1. The aircraft Local ID consists of a 2-bit prefix (Group ID) plus a 6-bit numerical suffix (Aircraft ID). The 6-bit numbers are used as follows:

0 Reserved for ground station or indication that the net is full1–60 Available for aircraft users61 Discrete addressing not supported/Dummy address62 Discrete addressing not supported (alternate validity window)/Dummy

addressspare63 Broadcast

3.4 M UPLINK FORMATS

3.4.1The M uplink messages consist of the synchronization sequence S2* and 96 bits. Of these, 48 are information bits, which are contained in four (24,12) Golay words. A message is created by assembling four separate words. There are six types of standard words listed below, along with their contents.

Beacon 1Message ID 4 bitsVoice Signal 2 bitsAircraft ID (Poll) 6 bits

Beacon 2Configuration 4 bitsSlot NumberBeacon ID 2 bitsGround Station Code 3 bitsSquelch Window 3 bits

Reservation

Local ID 8 bitsReserved Slot 4 bits

ICAO Address (2 words)ICAO Address 24 bits

Next Net 1Next Group ID 2 bitsNext Frequency 10 bits

Next Net 2Next Net Type 1 bitNext Ground Station Code 3 bitsLocal ID 8 bits

Terminate NetTerminated Local ID 8 bitsSpare 4 bits

- 4 –

Supported OptionsOptions 10 bitsRadio ID 2 bits

3.4.2 Beacon 1

3.4.2.1Every uplink M channel message begins with Beacon 1. The initial field of Beacon 1, the MID, identifies the type of information to follow. The first bit of all uplink MIDs is a 0. The remaining bits are described below:

MID 0000 Normal Message0001 Net Entry Response 1: No previous link0010 Net Entry Response 2: Previous link preserved0011 Next Net Command0100 Recovery010101100111

Handoff CheckTerminate NetSupported Options

10000110-1111

SpareReserved for Downlink Messages

Note.— There are two separate Net Entry Response messages. If the new ground station is unable to redirect the link from the previous ground station, MID = 0001 is used. If the new ground station can support redirection, MID = 0010 is used. The defined messages will be described more completely below.

3.4.2.3The final part of Beacon 1 is the Aircraft ID (Poll) field. This field is used to poll each of the members of a net in an orderly way. The Local ID suffix of the aircraft user to be polled is the 6-bit Aircraft ID field. For configurations other than 3T, 3S and 2S1X4V, 3V, 3V1D, the Group ID is the same as the Slot NumberBeacon ID in Beacon 2. For 3T, the Group ID is implied by the position of the Beacon 1 field in the uplink M channel burst. For 3S the Group ID is irrelevant and is arbitrarily set to A = 00. For the wide area coverage slots of 2S1X, the Group ID is also arbitrarily set to A = 00. For 1V3D and 1V2D, the Group ID provides 2-bit extension to the aircraft addressing for up to 240 users. For 2V2D and 2V1D, each user group has 2 Group IDs available to provide up to 120 users in each net. Polling is described in more detail in Chapter 4 on protocols.

3.4.3 Beacon 2

3.4.3.1Every uplink M channel message contains a Beacon 2 word that follows Beacon 1. Beacon 2 contains static information that is not expected to change from cycle to cycle. This information is required by radios entering a net. The first field, labelled Configuration, defines how the system is configured in the area covered by the ground transmitter. There can be up to 16 different configurations. Those that have been defined are as follows:

Configuration 0000 4V0001 3S0010 3V1D0011 2V2D0100 3T0101 3V0110 2V1D0111 2S1X

- 5 –

100010011010–1111

1V3D1V2DSpare

Note.— There can be different configurations associated with different ground stations at any given time. It is also possible to provide mixed configurations at a single ground site. For example, a single ground site could support a 2V2D net using slots A and C and two separate 4V nets using slots B and D.

3.4.3.2The field called Slot NumberBeacon ID is normally just the number of the slot in which the beacon is transmitted (A = 00, B = 01, C = 10, D = 11), except for 1V3D/1V2D where it is used for address extension up to 240 and 2V2D/2V1D where it is used to differentiate the 2 nets while doubling the address space to 120. The GSC is a 3-bit number described previously. The Squelch Window is a 3-bit number that is related to the physical size of the volume served by the ground station. More details on the Squelch Window will be found below.

3.4.4.3The exact meaning of the 4-bit Reserved Slot field is as follows:

0000 No reservation information0001 RACK (UDR Supported)0010 RACK (UDR Not Supported)Spare0011 Spare0100 Data reservation (start in even B slot)0101 Data reservation (start in odd B slot)0110 Data reservation (start in even B slot, skip D)0111 Data reservation (start in odd B slot, skip D)1000 Data reservation (start in even C slot)1001 Data reservation (start in odd C slot)1010 Data reservation (start in even C slot, skip D)1011 Data reservation (start in odd C slot, skip D)1100 Data reservation (start in even D slot)1101 Data reservation (start in odd D slot)1110 Voice reservation (start in even D slot)1111 Voice reservation (start in odd D slot)

Note.— For non-3T configurations, use only the codes with the third bit set to 0, except for 0010. For system configuration 2V2D, allowable values are 0000, 0001, 0010, 1000, 1001, 1100, and 1101. For system configuration 3V1D, allowable values are 0000, 0001, 0010, 1100, and 1101. For system configuration 2V1D, allowable values are 0000, 0001, 0010, 1000, and 1001. For configuration 1V3D the allowable values are 0000, 0001, 0010, 0100, 0101, 1000, 1001, 1100, and 1101. For configuration 1V2D the allowable values are 0000, 0001, 0010, 0100, 0101, 1000, and 1001.Only configuration 3T uses the codes 0100, 0101, 0110, 0111, 1010, 1011, 1110, and 1111.

3.4.7 Terminate net

Terminate Net is used by the ground station to reclaim a Local ID that has previously been assigned. This is primarily intended to be used when the net is nearly full and the ground station recognizes that some airframes have multiple radios in the net and hence are occupying multiple local IDs. By use of the Terminate Net message, the ground station can free up sufficient address space so that new aircraft entering the net can access discrete-addresses services.

3.4.8 Supported options

- 6 –

Supported Options is used to complete the net entry process and communicate whether certain options are actively being offered by the ground station. Currently, the support for Urgent Downlink Request is the only defined option. The second field of Supported Options is the Radio ID, which is used to differentiate different radio units on an airframe that may be tuned to the same net. As the DLS sublayer does not have access to the Local ID, the radio ID, passed in the Address Type field of the DLS, along with the ICAO address in the Aircraft Station Address field allows for the unique addressing of the DLS frames, and ensures that aircraft operating multiple radios on the net do not self-jam.

3.4.7 9 Message types

3.4.79.1This section shows how the various types of uplink messages are constructed.

Normal Message (MID = 0000)

Beacon 1 12 bitsBeacon 2 12 bitsReservation #1 12 bitsReservation #2 12 bits

Net Entry Response (MID = 0001 or 0010)

Beacon 1 12 bitsBeacon 2 12 bitsICAO Address 24 bits

Next Net (MID = 0011)

Beacon 1 12 bitsBeacon 2 12 bitsNext Net 1 12 bitsNext Net 2 12 bits

Recovery (MID = 0100)

Beacon 1 12 bitsBeacon 2 12 bitsReservation #1 12 bitsReservation #2 12 bits

Terminate Net

Beacon 1Beacon 2Terminate NetReservation

(MID = 0110)

12 bits12 bits12 bits12 bits

Supported Options

Beacon 1Beacon 2Reservation

(MID = 0111)

12 bits12 bits12 bits

- 7 –

Supported Options 12 bits

3.4.79.2The usage of the various uplink M channel message types will be explained in Chapter 4, Protocols.

3.5.3 The Poll Responses and Next Net ACKs consist of the synchronization sequence S2* and 48 bits. Of these 48 bits, 24 are information bits assembled as two (24, 12) Golay words. Included in each message is a 4-bit MID. To distinguish the Poll Responses and Next Net ACKs from the uplink M messages (which also use synchronization sequence S2*), the first bit of all such messages is a 1. The remaining bits are defined as follows:

MID 1000 4V/3V1D-Slot Configuration Poll Response (ALTS)1001 Spare2V2D Configuration Poll Response (ALT)1010 1V3D-Slot Configuration Poll Response (ALTS)1011 Spare3T Configuration Poll Response1100 3T 3V/3S/2S1X Configuration Poll Response (ALT)

1101 Spare2V1D Configuration Poll Response (ALT)1110 1V2D Configuration Poll Response (ALT)Spare1111 Next Net Acknowledgment

Note.— Poll Responses are Alternate Timing Signals (ALTS) except in the 3T configuration. The meaning of the designation ALTS will be explained in Chapter 8.

3.5.5 Message types

3.5.5.1The various downlink messages are defined below.

Poll Response (S2*: MID = 1000 , 1010,- 11010)Reservation Request (S1: MID = 1000)

Message ID 4 bitsLocal ID 8 bitsGround Station Code 3 bitsVoice Request 2 bitsNumber of Slots Required 4 bitsSpare 1 bitPriority 2 bits

Acknowledgment(S1: MID = 1001)

Message ID 4 bitsLocal ID 8 bitsGround Station Code 3 bitsVoice Request 2 bitsSpare 7 bits

Next Net ACK (S2*: MID = 1111)Leaving Net(S1: MID = 1010)

- 8 –

Message ID 4 bitsLocal ID 8 bitsGround Station Code 3 bitsAll zerosSpare 9 bits

Net Entry Request(S1*: No MID)

ICAO Address 24 bits

Note.— Every downlink message (except the Net Entry Request message) contains a MID, a Local ID and a GSC.

3.6 MEDIA ACCESS CONTROL MESSAGE FORMAT SUMMARY

The various message formats described in this chapter are summarized in Tables 3-1, 3-2 and 3-3. (The numbers following the colons represent the number of information bits contained.)

Table 3-1. Burst format summary (M channel uplink)

Type Message ID SyncNormal

Net Entry Response 1

Net Entry Response 2

Next Net

Recovery

Terminate Net

Supported Options

0000

0001

0010

0011

0100

0110

0111

S2*

S2*

S2*

S2*

S2*

S2*

S2*

{Beacon 1: 12}{Beacon 2: 12}{Res: 12}{Res: 12}

{Beacon 1: 12}{Beacon 2: 12}{ICAO Address: 24}

{Beacon 1: 12}{Beacon 2: 12}{ICAO Address: 24}

{Beacon 1: 12}{Beacon 2: 12}{Next Net 1: 12}{Next Net 2: 12}

{Beacon 1: 12}{Beacon 2: 12}{Res: 12}{Res: 12}

{Beacon 1: 12}{Beacon 2: 12}{Terminate: 12} {Res: 12}

{Beacon 1: 12}{Beacon 2: 12}{Res: 12}{Opt: 12}

{ } = Golay Codeword (two words in the case of ICAO address)

{Beacon 1: 12} = {(Message ID: 4)(Voice Signal: 2)(Aircraft ID (Poll): 6)}

{Beacon 2: 12} = {(Configuration: 4)(Slot NumberBeacon ID: 2)(Ground Station Code: 3)(Squelch Window: 3)}

{Next Net 1: 12} = {(Next Group ID: 2)(Next Frequency: 10)}

{Next Net 2: 12} = {(Next Net Type: 1)(Next Ground Station Code: 3)(Local ID: 8)}

{Reservation: 12} = {(Local ID: 8)(Reserved Slot: 4)}

{Terminate: 12} = {(Local ID: 8)(Spare: 4)}

{Options: 12} = {(Options:10)(Radio ID: 2)}

- 9 –

Table 3-2. Burst format summary (M channel downlink)

Type Message ID Sync

Res. Request

Acknowledgment

Leaving Net

Poll Response

Next Net ACK

Net Entry Request

1000

1001

1010

1000/1010/1100

1111

N/A

S1

S1

S1

S2*

S2*

S1*

{(Message ID:4)(Local ID: 8)}{(GSC: 3) (Voice Request: 2) (Number of Slots: 4) (Spare: 1)(Priority: 2)}

{(Message ID:4)(Local ID: 8)}{(GSC: 3) (Voice Request: 2)(Spare: 7)}

{(Message ID:4)(Local ID: 8)}{(GSC: 3)(SpareAll 0: 9)}

{(Message ID:4)(Local ID: 8)}{(GSC: 3) (Voice Request: 2)(Number of Slots: 4) (Spare: 1)(Priority: 2)}

{(Message ID:4)(Local ID: 8)}{(GSC: 3)(Spare: 9)}

{ICAO Address: 24}

3.7.2 The Address Type field indicates the direction of communication as well as indicating whether the address is a point-to-point address or a multicast address. The Address Type also conveys the Radio ID to differentiate multiple radios on a single airframe. The field is encoded as follows:

Address Type 000 Directed Ground-to-Air (Radio ID=0)

001 Directed Air-to-Ground (Radio ID=0)

010 Directed Ground-to-Air (Radio ID=1)


100 Directed Ground-to-Air (Radio ID=2)


010-110 Reserved

111 Broadcast Ground-to-Air

3.7.3 The Frame Type field indicates the use of the frame contents. It is encoded as follows:

Encoding Frame Type

p0p10 Information (INFO),

where p0p1 represents the priority encoding

001 Control-Command (CTRL_CMD)

011 Control-Response (CTRL_RSP)

101 spare

111 Acknowledge (ACK)

3.9.3 Raw interface

The raw interface is identified by a Network Type field of 2. The Compression Type (CT) field is always set to 0. Immediately following the payload byte is an 8-bit Initial Protocol Identifier (IPI), encoded per ISO 9577. For an IPI of 0xFF, an 8-bit Extended Initial Protocol Identifier (EIPI) follows. An EIPI of 0xFF is reserved to delineate ACARS networking over VDL Mode 3. An EIPI of 0xFE is reserved to delineate FIS-B messaging consistent with RTCA DO-267A. This interface is used to imply no networking protocol is being usedAn EIPI of 0 is reserved . This is primarily intended to support Minimum Operational Performance Standards (MOPS) testing. It This interface could also be used for a dedicated link, if so desired by a given State.

3.10 VOCODER FORMATTING

3.10.1 Bit Exact Algorithm

A proprietary vocoder algorithm developed by DVSI (Digital Voice Systems Incorporated) has been selected for use by the VDL Mode 3 system. DVSI has developed a fixed-point C implementation of the selected vocoder algorithm and a corresponding set of test vectors. The fixed-point C application will be ported by DVSI to processor-specific versions of code that can be delivered to implementers, or implementers can utilize the C reference for their own port. Each manufacturer will then have to integrate the ported code into its product. For certification purposes, implementers will wish to use the bit-exact fixed-point implementation available from DVSI. This implementation provides repeatability of output necessary to satisfy certification of the implementation.

3.10.2 Bit Ordering

Implementers should be aware that the bit ordering specified in Part II, section 8.2.3.1 is nibblewise bit-reversed (every 4 bits reversed) from what DVSI indicates in their reference documentation.

3.10.3 Test Vectors

The DVSI part number for the test vectors and output results is 03-168-1229 rev 1.0 as of October 2003.

Part I – Implementation aspectsChapter 4 – Protocols

4.2 Link establishment

All aircraft radios perform link establishment within a particular net prior to performing any other function. A net is determined by a particular frequency, Group ID (i.e., the channel label suffix), and GSC.

4.2.1.3If the entrant identifies one or more signals with the correct GSC, it will then check the Configuration field. If the configuration is not 3T or 3S4V, 3V1D, 3V, it will then check that the Slot NumberBeacon ID field agrees with the Group ID of the net it is trying to enter. (If the configuration is 3T, the Group ID and Slot NumberBeacon ID are always A may not agree, and the special procedures necessary are described in this Chapter . For 3S/2S1X, the entrant will choose whichever slot (A, B or C) appears to provide the “best” signal as described in Chapter 5. For 2V2D/2V1D, there are 2 viable Beacon IDs applicable to each user group for expanded addressing. For 1V3D/1V2D, Beacon ID is used for extended address of the single user group.) Having identified a candidate beacon signal, the aircraft radio will verify it by detecting the same beacon in a later MAC cycle and determining that the information in Beacon 2 has not changed (except for a possible change to a different compatible Beacon ID). (In general, a beacon is considered valid only if the information in the Beacon 2 word is the same as the information in the Beacon 2 word in the previous beacon reception. The effect of this is that a minimum of two beacon receptions is required for net initialization.) If all the tests are passed, the new net’s configuration and timing are known, and the aircraft radio enters Timing State 1 (TS1). It then can proceed to the next stage—net entry.

4.2.2 Net entry

4.2.2.1A radio can, as an option, begin operation immediately after net initialization, i.e. as soon as it enters TS1. However, it will be able to transmit only voice messages and will not be able to use functions requiring downlink M channels, such as two-way data transmission. Normally, a net sign-in procedure must occur. (Note that a ground station beacon with Local Aircraft ID = 61 indicates that the ground station does not support discrete addressing, and net entry should not be attempted.) This process is begun when the aircraft radio transmits a Net Entry Request message in a downlink M channel RA opportunity. Prior to signing in, the entrant must determine the location of the possible RA opportunities. This is done by listening to the net for one entire cycle. During this listening period, the entrant notes which slots are scheduled for acknowledgments (ACKs) and poll responses. This process can take place in the same MAC cycle during which the beacon was “verified.”

4.2.2.5The pool of Local ID numbers that are available to be distributed depends on the configuration of the net. For instance, for all configurations except 3T and 3S4V, 3V1D, 3V the Local ID prefix always agrees with the User Group ID. However, for the 3T configuration the Local ID prefix is chosen by the ground station based on other considerations, as described in Chapter 6. For 3S/2S1X the Group ID is not relevant, and a single, arbitrary ID (i.e. A) is used for all users. For 1V3D/1V2D, the User Group ID is always A, but the Local ID prefix can be any from A-D. For 2V2D/1V2D, the Local ID prefix for User Group A can be A or C, while the Local ID prefix for User Group B can be B or D.


- 15 –

4.2.2.6There can be up to 60 Aircraft IDs provided for each Local ID prefix. The ground station will choose an ID number between 1 and 60. Number 0 is reserved for the ground station. Numbers will be assigned to the aircraft users in sequence. Thus, if the previous entrant had been given number M, then the new entrant will be assigned number M+1. If the number M+1 has already been assigned then the next higher available number will be chosen. For the purposes of this algorithm, the numbers are “rolled over” so that the next number after 60 is 1. For system configurations 2V2D and 2V1D, the User Group ID will change once to provide an effective 120 address space. For system configurations 1V3D and 1V2D, the User Group ID will change once to provide an effective 240 address space. If all the numbers are used up, the number 0 will be given to indicate a full net. If Aircraft ID = 0 is indicated, the aircraft radio can still participate in the net in a limited way (using the dummy ID = 61). In this case, the aircraft radio becomes the equivalent of a radio that has not “signed in,” which cannot provide features that require discrete addressing. The aircraft user should be given some indication (to be determined) that advises when it is entering the net with reduced capability. If this is not acceptable the operator can take appropriate action.

4.2.2.8If the Poll Response has the correct information (i.e. the entrant echoes back the correct Local ID) then the ground radio will confirm that the net entry procedure has been successfully completed by sending some sort of Reservation message Response to the entrant in a Supported Options message. If the entrant requested slots in its Poll Response and slots are available in the next cycle, an actual reservation will be delivered. For all other cases, a RACK must be delivered. Note that this is the only case where a RACK is delivered after a poll response. Otherwise, RACKs are provided only for RA Request Messages. The Supported Options message includes the Radio Identifier assigned to the aircraft as well as an indication of whether the ground station supports any of the optional capabilities, such as Urgent Downlink Request.

4.2.3 Net entry retransmission

4.2.3.2To account for a possible delayed response the net entrant will wait for a number of MAC cycles equal to WE before it initiates another Net Entry Request using a new random number as discussed in 4.11. This process may result in a duplicate Net Entry Request (i.e. with the same ICAO address) being received by the ground radio before it can respond. If this occurs, the second request will be rejectedreceive a different Local ID and Radio ID. The aircraft radio will receive one of the responses and ignore the other. The extra Local ID will be recovered when TL4 expires due to lack of Poll Responses.

4.2.4 Multiple radio operation

4.2.4.1Some aircraft operate multiple VHF ATC communications radios concurrently. To ensure proper operation of the VDL Mode 3 protocols, the ground station assigns a Radio ID in addition to the Local ID when a radio enters a net. As data messaging does not include the Local ID, the ICAO address alone is insufficient to differentiate the multiple units that may be tuned to a net (or same frequency for 3V1D/2V1D). If they are not differentiated, multiple radios might attempt to acknowledge uplinks and respond to polls. To prevent this, different Local IDs are assigned to subsequent Net Entry Request messages from the same ICAO address along with an incremented Radio ID. The Radio ID allows all of the stations to differentiate the different radios so that only a single unit acknowledges uplink data transmissions.

4.2.4.2The downside of this solution is that more addresses will be used when aircraft operate multiple radios in the same frequency/net. To help mitigate this issue, the ground station has a means to recover Local IDs from aircraft which may be holding multiple IDs when it nears maximum capacity in the net. Also, 2V2D and 2V1D system configurations make use of the B/D beacon IDs to expand the net Local ID space to 120.

Section 4.2.3.3 deleted4.2.3.3Each time a Net Entry Request fails to elicit a response, a counter is incremented. If this counter reaches the threshold, NL1, an upper layer protocol (SN-SME) is notified and the data link connection (if any) may be declared broken.

- 16 –

4.3.4The third type of window is the validity window. The validity window is used to control the processing of M channel messages. When an aircraft radio is in net initialization mode (TS0) its validity window is wide open; it is looking for any M channel signal. After net initialization, when it is searching for uplink M channel signals, its validity window should be reduced to the range from -1 K(n+1)+1 to +1 symbol periods with respect to the timing of its most recent successful beacon reception, where K=4 for 4-slot configurations and K=8 for 3-slot configurations and n=squelch window parameter. (The finite size of the range is to account for clock drift between M channel receptions; it is assumed that the aircraft radio will update its clock often enough to keep its error significantly less than one symbol period under normal circumstances.). This allows the aircraft radio to track the closest ground station. Due to a failure or other circumstance, a particular ground station may be removed and/or replaced. In this case, the distance between an aircraft radio and its nearest ground station might increase abruptly. To accommodate this possibility, the aircraft needs to react in a timely fashion. To allow it to do so, it should extend its validity window to include all relative time differences whenever its counter CTC1 exceeds 24. This change will enable it to find a valid beacon before it enters timing state TS2. A beacon is considered valid if the information in the Beacon 2 word (including GSC) is unchanged from the previous Beacon 2 information and if it falls within the current validity window. If either of these conditions is not met, the new Beacon 2 information and timing information will be held in escrow until it is confirmed by the next beacon reception. Note that this beacon processing rule also applies to cases where the ground timing shifts abruptly (perhaps due to a switch from main to backup transmitters) as described in Chapter 9, 9.6.

4.3.5 A beacon is considered valid if the information in the Beacon 2 word (including GSC) is unchanged from the previous Beacon 2 information (except for a possible change to a different compatible Beacon ID) and if it falls within the current validity window. If either of these conditions is not met, the new Beacon 2 information and timing information will be held in escrow until it is confirmed by the next beacon reception. However, the aircraft radio needs to process any signaling commands contained in uplink beacons that have the appropriate GSC and Beacon ID(s) regardless of the validity window. Note that this beacon processing rule also applies to cases where the ground timing shifts abruptly (perhaps due to a switch from main to backup transmitters) as described in Chapter 9, 9.6.In cases where the ground station is supporting a select-key operation, as indicated by aircraft ID 62 in the ground beacon, the validity window should be increased to the range from –K(n+1)+1 to +1, where, K=4 for 4-slot configurations and K=8 for 3-slot configurations, and n is the squelch window parameter. (Beacons outside this window should be completely ignored and not held in escrow, as above.) This allows the aircraft radio to track the closest ground station. Due to a failure or other circumstance, a particular ground station may be removed and/or replaced. In this case, the distance between an aircraft radio and its nearest ground station might increase abruptly. To accommodate this possibility, the aircraft needs to react in a timely fashion. To allow it to do so, it should extend its validity window to include all relative time differences whenever its counter CTC1 exceeds 24. This change will enable it to find a valid beacon before it enters timing state TS2.

4.4 POLLING

4.4.1In order to provide guaranteed access for reservation requests and link management functions, the slotted aloha reservation protocol provided by the RA downlink M channel slots is supplemented by an orderly polling procedure. Every uplink M channel message except the Handoff Check message can support a polling command from the ground station using the Beacon 1 word. In the absence of a net entrant, the ground station will sequentially poll all of its “registered” aircraft radios in ascending numerical order. (The numbers are “rolled over” so that 1 follows 60 or whatever the highest ID is.)

4.4.3 As stated in 4.4.1, the normal polling sequence will be in ascending numerical order. The schedule depends on the configuration and on the number of users as follows. For configurations 2V2D and 3T, if the current number of users with a particular Group ID is N, then:

a) If N 4, there will be one poll command every fourth cycle.b) If 4 < N 8, there will be a poll command every other cycle.c) If 8 < N 12, there will be one “no poll” every fourth cycle.d) If 12 < N, there will be a poll command every cycle.

- 17 –

4.4.4These rules guarantee that (in the absence of net entrants) every user is polled at least as often as once per 3.84 seconds, wherever there are 16 or fewer users. The “no poll” message will be denoted by the value 0 in the Aircraft ID (Poll) field. This feature will provide opportunities for net entrants to use the poll response slot periodically for net entry requests.

4.4.5For configuration other than 2V2D and 3T the above sequence ends at the second step. Thus, the maximum polling frequency is every other cycle.

4.4.6Note that if a ground station does not support polling or any other type of downlink M channel activity, it will transmit the value 61 in the Aircraft ID (Poll) field of its uplink M channel message.

4.4.74Below is a summary of meanings of the various possible entries in the Aircraft ID (Poll) field of Beacon 1.

0 In the case of a Net Entry message, this means that the net supports addressing, but there are no available addresses at this time. For all other message types, this means that no poll response is scheduled in the next cycle. In either case, the poll response slot in the next cycle is available for RA.

1–60 In the case of a Net Entry message, this is the new ID number of the entrant identified by the attached ICAO number. In any case, this is the ID of the aircraft user who must respond in the poll response slot of the next cycle.

61 This means that the ground station does not support addressing. Only voice communication is possible (and possibly uplink data broadcasts).

4.4.5 Polling algorithms

4.4.5.1 Background

4.4.5.1.1This section proposes a means of minimizing potential interference caused by Poll Responses transmitted by a VDL Mode 3 radio. In aircraft cosite situations, VDL Mode 3 Poll Responses transmitted at a high rate might cause annoying and/or unacceptable interference to VHF radios operating on other frequencies (possibly using VDL Mode 2 or double side band AM). Note that in many cases the integrated voice and data capability of VDL Mode 3 should make it unnecessary to configure aircraft communications in such a “mixed-mode” way; nevertheless, the situation may arise. Interference can be mitigated by scheduling the polls so that an individual aircraft radio is not polled more often than once per six seconds. This is a requirement on the ground system, which determines the polling schedule.

4.4.5.1.1To implement the six-second rule while providing all the other services demanded of the up link beacon signals requires a nontrivial algorithm, particularly in site diversity sectors. A possible single site algorithm, describing a simple implementation, and providing an example of the algorithm operating in a particular environment, follows. Also, a final section describes an extension of the algorithm applicable to site diversity sectors.

4.4.5.2 Poll Scheduling Algorithm

4.4.5.2.1 The scheme can be defined as follows:

For each user (labeled k) in the net, there is an associated Poll Counter (PCk). As each user enters the net, it is assigned a PC value of –25 immediately after its Net Entry Response is sent. At the beginning of each MAC cycle, every user’s counter is incremented by (plus) one. At the middle of each MAC cycle, the ground station polls the user with the highest nonnegative PC value. Whenever a user is polled, its counter is reset to –25.

Net Entry Responses have precedence over ordinary polls. If a Net Entry Response is required during a MAC cycle when a poll would otherwise take place, the Net Entry Response will occur instead.

- 18 –

4.4.5.2.2This algorithm prevents users from being polled more often than once per 6 seconds because it takes 25 MAC cycles for a counter to change from –25 to 0. It also has the property that it polls the user that has waited the longest since its last poll. When there are few users in a net, this algorithm will result in each user being polled almost exactly once per 6 seconds. (This could be increased slightly if the ground station is involved in net entry at the time when a poll would ordinarily happen.) As the number of users increases, the algorithm will automatically adjust the average interval to more than 6 seconds.

4.4.5.2.3This technique will work for all VDL Mode 3 configurations, except for a very slight modification for 4V, 3V, 3V1D, 2V1D, 3S and 2S1X. For those configurations only, a poll should be sent on at most half the MAC cycles (in order to allow opportunities for down link net entry bursts for the voice-only system configuration or in the case of 3V1D/2V1D to ensure the ground station can properly associate the appropriate user group to the net entry request). This would be easy to implement by requiring that polls only take place in “even numbered” MAC cycles (or by not allowing polls in consecutive MAC cycles).

4.4.5.3 Implementation

4.4.5.3.1The algorithm can be easily implemented in the ground station by creating an ordered list containing each user’s Local ID and its associated value of PCk (the Local ID and k are essentially synonymous). The list is sorted by PCk, with the highest value on top. At the beginning of each MAC cycle all PC values are increased by one. If that operation results in the PC of the top entry in the table being nonnegative, it should be polled (unless a Net Entry Response is pending). If the Local ID at the top of the list is polled, it is immediately moved to the bottom of the list and its PC is set to –25. If, instead, the beacon is used to service an entrant, the Local ID of the entrant is immediately added to the bottom of the list, along with the PC value of –25. This procedure ensures that no two entries in the list have the same PC value and that the list remains ordered.

4.4.5.3.2The use of PCk also works for site diversity sectors, with the proviso that each ground station has a separate list (see Section 4.4.5.5). An added complication is that when an airborne user is shifted from one ground station to another, it must be placed in the new table in the correct position.

4.4.5.4 Performance

4.4.5.4.1The performance of this algorithm was investigated using a computer simulation. Figure 4-1a shows the results of a particular simulation run for a 2V2D-type configuration in which polling can occur in every cycle. These results would be similar for System Configurations 1V3D and 1V2D Figure 4-1-b shows the results of a similar simulation run for a 4V-type configuration in which polling is normally allowed only in “even-numbered” cycles . These results would be similar for System Configurations 3V, 3V1D, 2V1D, 3S and 2S1X. Aircraft were allowed to randomly enter and leave a net in such a way that the number of participants increased from zero, reached a plateau, fell below 25, and then reached another plateau. The rates at which the entering and leaving events were allowed to happen were artificially high in order to stress the algorithm. The curves represent the average polling interval per user (averaged over contiguous 12 second sample periods) and the number of users present (at the end of each 12 second sample period). The important features of the figure are that:

(1) The interval between polls (of each user) is never less than 6 seconds, and

(2) The interval increases just enough to accommodate the number of users in the net. (For 2V2D, the minimum time is approximately equal to the number of users times the duration of a MAC cycle, i.e., 0.24 seconds. For 4V, the minimum time is approximately equal to the number of users times the duration of two MAC cycles, i.e., 0.48 seconds)

- 19 –

Polling Algorithm

01.22.43.64.8

67.28.49.6

10.812

0 4 8 12 16

Time (Minutes)

Inte

rval

(Sec

onds

)

0

10

20

30

40

50

Num

ber o

f Use

rs

Average poll interval

Number of users

Figure 4-1a: Performance of the Poll Scheduling Algorithm (2V2D, 1V3D, 1V2D)

Polling Algorithm

0

6

12

18

24

30

0 4 8 12 16

Time (Minutes)

Inte

rval

(Sec

onds

)

0

12.5

25

37.5

50

62.5

Num

ber o

f Use

rsAverage poll intervalNumber of users

Figure 4-2b: Performance of the Poll Scheduling Algorithm (4V, 3V, 3V1D, 2V1D, 3S, 2S1X)

4.4.5.5 Enhanced algorithm for site diversity

4.4.5.5.1The algorithm described thus far could be used (in a slightly modified form) in diversity site environments. The simplest way to use it would be to allow beacon pairs from the ground sites to be shifted from site to site more or less periodically. As each site is used to transmit a beacon, the beacon signal can also be used to poll the user attached to that site that has the highest nonnegative PC (if any). This approach would work well in an environment with nearly equal numbers of users associated with each ground site. However, if some sites have many users and some have few, the beacons in the underpopulated sites will often not be used for polls. Also, the users in the overpopulated sites will experience long times between being polled. This last effect may artificially limit the number of users that can be supported in a given area.

4.4.5.5.2There are other issues associated with site diversity and beacon usage. First, during net entry the beacon must remain on a single site for three cycles. This is because one cycle is used for the Net Entry Response, the next

- 20 –

cycle anticipates the initial Poll Response from the entrant, and the third cycle is needed for the Reservation or RACK that must be sent to the entrant in order to complete the entry process. Second, provision must be made to send up link Next Net messages in a timely manner. The next net information occupies a different part of the up link message from the polling information, but the fact that the recipient of the Next Net message is “connected” to a particular ground station influences the choice of ground site in the diversity case. Similarly, the choice of ground site may be influenced by the need to send a Reservation to a particular user. Finally, the desire to deliver ground voice preemption to an active aircraft radio may influence the choice of ground site for a particular beacon transmission.

4.4.5.5.3These issues can be addressed using a more sophisticated algorithm. The proposed algorithm for diversity sites attempts to meet the following goals:

1. Poll each user no more often than once per six seconds.

2. Transmit a beacon pair from each ground site as often as possible, at least one pair each six seconds. (Beacon pairs, i.e., two beacons from the same site in consecutive MAC cycles, are necessary to support net initialization. Also, beacons sent at six-second intervals provide each aircraft two independent chances of receiving a beacon before entering Timing State 2 (TS2).)

3. Allow Net Entry Responses to be transmitted as quickly as possible.

4. Allow overpopulated ground sites to transmit polls at proportionately higher rates.

5. Address the requirements to send voice preemption, next net, and data reservation information to particular users in a timely fashion.

4.4.5.5.4These goals can be met through the use of two counters in addition to the previously defined Poll Counter (PCk). The first new counter is the Beacon Counter (BCj), where the index j runs from 1 to N (N being the number of ground sites). BCj counts the number of MAC cycles that site j has waited since transmitting its last beacon. The second new counter is the Current Beacon Counter (CC). CC, which is defined below, is used to ensure that the number of consecutive beacons transmitted from any site is at least two.

4.4.5.5.5Like PC, BC is incremented by 1 at the beginning of each MAC cycle. BC j is reset to 0 whenever site j transmits a beacon. (Note that when the ground system is initialized, the BC values of the ground sites can be set to any positive values so long as no two ground sites have the same value.) Whenever the values of BC are all small, the number of users attached to each site is allowed to dictate the beacon cycle. However, if any BC is at or above some threshold K (tentatively set to 22) the need to send a beacon pair takes precedence.

4.4.5.5.6To allow timely net entry, an up link Net Entry Response and the subsequent Reservation Response or RACK are allowed to take precedence over routine polling, except in cases where sending a Net Entry Response would result in transmitting an isolated beacon (i.e., a single beacon from an individual ground site). The CC process enforces this requirement:

4.4.5.5.7To describe the algorithm in detail, the rules are presented graphically in Figures 4-2 through 4-4. Figure 4-2 shows how CC is incremented each MAC cycle. When a beacon is transmitted, the state moves to the right. If the beacon site is changed, CC is initially reset to 0 and then incremented by 1 when the beacon is sent. The point of CC is that the site of the beacon transmission is not allowed to be changed when CC=1.

- 21 –

0 1 2

newsite

CC State Diagram

samesite

samesite same

site

Figure 4-3: Rules for Updating CC

4.4.5.5.8 Figure 4-3 shows how the counters are used to determine which site gets to transmit the beacon for a particular MAC cycle. The site selection algorithm gives precedence to potential actions in the following order:

1. Complete a beacon pair2. Service a pending net entry procedure3. Service a site with BC above threshold4. Service a user subject to voice preemption5. Service the recipient of a Next Net Message6. Service a user with nonnegative PC7. Service the recipient of a data Reservation or RACK8. Service the site with the longest wait to send a beacon

4.4.5.5.9 When a site is chosen, the user that is polled in the beacon signal is chosen (from among the users attached to the site) according to the following priority order:

1. Pending net entrant2. Highest nonnegative PC3. None (if there are no nonnegative PC values)

4.4.5.5.10In either of the first two cases, the PC associated with the user receiving the Net Entry Response or poll is reset to –25 after the message is sent.

- 22 –

CC=1 ?

PendingNE Process ?

Any site withBC>=K ?

Any site withPC>=0 ?

Current site retains beacon

Site with entrant gets beacon*

Site with highest BC gets beacon

Site with highest PC gets beacon

Site with highest BC gets beacon

yes

yes

yes

yes

no

no

no

no

* Site chosen may be current site

Voice Preemption?

Site with aircraft voice user gets beacon*yes

no

Next NetPending?

Site with Next Net recipient gets beacon*yes

no

Reservation orRACK pending?

Site with Reservation recipient gets beacon*yes

no

Figure 4-4: Site selection algorithm

4.4.5.5.11The operation of the algorithm for each MAC cycle is summarized in Figure 4-4. This algorithm may seem to be complicated, but it is just a series of very simple individual steps. It can be implemented with very little software and memory.

- 23 –

Increment all BC j and PCkBeginning ofMAC cycle

End ofMAC cycle

TransmitBeacon

Choose beacon site usingSite Selection Algorithm.

If new site chosen, set CC=0.

Choose user to be polled, if any.

Reset BC of beacon site to 0.

Reset PC of polled user (if any) to -25.

Update CC using CC State Diagram.

Figure 4-5: Beacon/Polling Timing Diagram

4.4.5.5.12It may be instructive to observe how the algorithm would behave in certain conditions. Suppose for example that there are 2 users attached to each of 6 ground sites and that there are no current requirements for preemption, Next Net transmission, etc. In that case the beacon pairs will simply rotate in a regular fashion among the sites with a period of 2.88 seconds (= 12 x 0.24 seconds). About half the beacons will not contain polls because the minimum poll time is 6 seconds. If a new aircraft initiates net entry, the nearest ground site will send a Net Entry Response in the earliest possible MAC cycle that does not involve breaking up a beacon pair. This will result in a realignment of the beacon cycle since the servicing of a net entrant involves three cycles: the Net Entry Response, the initial Poll Response, and the subsequent up link Reservation or RACK.

4.4.5.5.13Next, suppose that one of the 6 ground sites has 30 users and the other sites have none. In that case, most beacons will be transmitted from the overpopulated site, with an occasional pair coming from one of the other sites. The polling cycle time will be approximately 12 seconds (= (30 + 20) x 0.24 seconds). Note that if the simplistic algorithm described at the beginning of this section had been used in this case, the approximate polling cycle time would have been 43.2 seconds (= 15 x 2.88 seconds). Thus, the proposed algorithm can greatly reduce the potential penalty that must be paid to provide periodic beacons throughout a large service volume.

4.4.5.5.14An interesting issue that arises with the use of this algorithm, or any other algorithm that does not give the delivery of data reservations the highest priority, concerns the scheduling of down link data slots. The Reservation message can only be delivered to a particular aircraft when a beacon is being transmitted through the ground site to which the aircraft is connected. This means that ground system needs to maintain separate queues for aircraft connected to each ground site so that it can provide a reservation to the user (if any) at the top of the queue associated with the current beacon. This adds some additional complexity, and it can lead to some inefficiency since users in other queues may have to wait extra cycles for their reservations.

4.4.5.5.15Note that this algorithm should be considered as an example of a possible method for beacon site selection. It is subject to replacement or change. For example, the order of priority that is implied by Figure 4-3 could be amended to optimize overall performance.

4.5 NET EXIT

- 24 –

The process of leaving a net will be enabled manually by the aircraft user of a radio that is currently signed into a particular net. (This process can also be automatic in the 3T mode.) The identity of the new net can be entered manually or automatically (via the Next Net message).

4.5.3.3Note that for the ground system to accept a Leaving Net message, it must verify the trailing zeros in the message are all zero. If the bits are not zero, then the ground station will ignore the message. If it was a valid Leaving Net message that contained uncorrected bit errors, the ground station will eventually drop the aircraft when NL2 expires.

4.6 AIRCRAFT VOICE TRANSMISSION (NOT 3T)

4.6.1Under normal circumstances, an aircraft user waits for a quiet period and begins voice communication by activating push to talk (PTT). This will cause the radio to begin transmitting in the first available voice time slot. The header will contain MID = 001, the Local ID, and EOM = 0. A radio that has no local ID will substitute the dummy ID = 61 for its Aircraft ID. It may happen that the radio will begin transmitting before a full complement of vocoder frames is available (there may even be no vocoder frames to transmit). If this occurs, the unfilled frames should be filled with a “silence” pattern. When PTT has been deactivated, the header will contain EOM = 1 in the last time slot of the V/D (voice) burst sequence. Note that the maximum duration of downlink voice message is 35 seconds. If PTT is activated for a longer time, transmission will cease and the user will be notified.

4.6.2A net entrant can transmit a voice message prior to completing the net entry procedure. However, such a radio must use the dummy Local Aircraft ID value (= 61) and the Group ID associated with the channel to which it is tuned until its new Local ID is confirmed by an appropriate Reservation message.

4.6.3Upon receipt of a voice message, the ground radio will change the beacon message Voice Access field from 10 (idle) to 01 (occupied).

4.6.6If the aircraft user has an urgent message, a “priority”n Urgent Downlink Request (UDR) M channel signal can be sent. The Voice Request field will read 10. When the ground radio receives this message, it can activate a special feature on the ground display that will indicate an urgent/priority message. The controller can then take appropriate measures. This feature could be supported as a future option. The ground station indicates whether or not it supports this capability in the Supported Options message sent upon an aircraft completing Net Entry.

4.6.7The slots used for sending and retransmitting the urgent/priorityUDR message are chosen using the same rules that are used for making a data reservation request. The aircraft user will continue sending the message until it receives a Reservation Response or a RACK.

Note.— An EOM normally appears in the header of the last burst of a voice message. For a number of reasons, it may happen that this EOM is not received. In this case, the receiving radio will continue to “look for” the absent voice signal. In order to account for this possibility, the radio should declare the message ended if two successive bursts are missed. During this time, the receiving vocoder should be supplied an appropriate message. This procedure allows the receiver to ride through a short fade while still purging vanished signals in a timely way. For the purposes of this procedure, missing a synchronization or receiving a header that fails validation constitutes a missed burst.

4.7.4It is clear from the previous paragraph that uplink voice messages will can pre-empt downlink voice messages, if this feature is enabled. Even if an aircraft user has activated PTT, its radio will cease transmitting. The aircraft user will be aware of this change because the uplink voice message will be heard. This pre-emption feature will allow the controller to correct the “stuck microphone” problem verbally.

- 25 –

4.8.4It may happen that a reservation request is not received by the ground station. This can happen due to a conflict or a noisy channel. Also, it is possible that a Reservation or a RACK cannot be transmitted in a timely fashion. In either case, the aircraft radio can retransmit its request after waiting WR MAC cycles. The exact retransmission algorithm is described in 4.11. To reduce the possibility of reservation conflicts, the retransmitted request must be for a number of slots equal to or less than the number requested previously. This is necessary because the eventual Reservation message cannot indicate which request elicited it. To increase the number of slots requested, an aircraft user must wait until its next poll response. The RA process may continue until the aircraft radio receives a poll command, after which it can continue repeating the request only in each subsequent Poll Response. The aircraft radio can transmit up to NM1 (default =20) Reservation Requests for a particular data message. If NM1 is exceeded, reservation request retransmission ceases and a higher level protocol (SN-SME) is notified.

4.10.2The ground station informs a user that its Reservation Request has been honoured by transmitting a Reservation addressed to the specified user indicating to it when in the next cycle it should begin transmitting. The aircraft user will begin its data transmission on that slot, and it will use all consecutive available data slots until it has no more data to send or until it has used as many slots as it requested in its last Reservation Request, whichever is less. (Note that under most circumstances the number of slots an aircraft radio will use is the same as the number it asked for, not less. However, an aircraft radio would, in principle, be allowed to substitute a higher priority message if it were no longer than the lower priority message that elicited the original request, and it has not made a previous attempt to transmit the frame group.) The scheduling algorithm provides for sequential access to the available slots. One user’s data reservation is completed before another has access.

4.10.8 Resolving reservation scheduling conflicts

4.10.8.1The uplink M channel is used to deliver many types of messages including Net Entry Responses, Next Net messages, Reservations and RACKs. It can happen that in a particular MAC cycle there are more demands on the uplink M channel than it can satisfy. To deal with such situations, the use of the channel is governed by the following priority ordering (from highest to lowest):

(1) Net Entry Response message

(2) Supported Options message with Reservation or RACK to confirm successful net entry

(3) Terminate Net message

(4) Next Net message

(54) Normal message with Reservation

(65) Normal message with RACK

4.11.5When an aircraft radio determines that it needs to transmit a RA message, it will send the message on the next available RA opportunity. For the Leaving Net message, this will be the only transmission; however, the Net Entry Request and the Reservation Request may be retransmitted if a satisfactory response is not received in a timely fashion. An aircraft radio will wait WE and WR MAC cycles before reinitiating the random transmission process for Net Entry Requests and Reservation Requests, respectively. If it determines it needs to retransmit the RA message, it will choose a random number, r, such that

0 ≤ r ≤ Rn

where n = E, R, L depending on whether the slot is being chosen for a Net Entry Request, a Reservation Request, or a Leaving Net message, respectively. The number r is chosen according to the pseudorandom process described in Chapter 9, 9.1. The aircraft radio will skip the next r available slots from the reduced list of allowed slots and transmit

- 26 –

in the next one. For the Leaving Net message, this will be the only transmission; however, the Net Entry Request and the Reservation Request may be retransmitted if a satisfactory response is not received in a timely fashion. An aircraft radio will wait WE and WR MAC cycles before reinitiating the random transmission process for Net Entry Requests and Reservation Requests, respectively. These processes will repeat indefinitely until they are terminated by a valid response or a poll (in the case of a Reservation Request), or the processes exceed time limits, i.e. thresholds NL1 (see 4.2.3) or NM1 (see 4.8) are reached. Rn and Wn can depend on the system configuration. Default values are listed in Table 4-2.

4.11.6This RA process is similar to a standard p-persistence/back-off procedure, but it appears to be easier to implement and does a better job of “balancing” the usage of the various slots in a MAC cycle.

Table 4-2. Random access parameters

Configuration4V 3V1D 2V2D 3T 3V 2V1D 3S 2S1X 1V3D 1V2D

RL03 08 08 08 03 08 03 03

0 0RE 3 8 8 8 3 8 3 3 8 8RR 3 8 8 8 3 8 3 3 8 8WE 1 2 2 2 1 2 1 1 4 4WR 1 2 2 2 1 2 1 1 2 2

4.12 DATA ERROR DETECTION AND CORRECTION

All data messages are protected by error correction algorithms on a slot-by-slot basis. The error correction is provided by an RS (72, 62) code that operates on 8-bit (28-ary) symbols and corrects up to five symbol errors. Error detection is provided by a 24-bit CRC, which is part of the DLS sub-layer (Chapter 5) protocol. Any message segment that passes the error detection process is considered to be correct. Information on the performance of these codes can be found in item 4 in the List of References to Part I of this manual. Where applicable, initial register values are assumed initialized to zero. For the CRC, the registers are reflected and the final register is not XORed.

4.14.5The following hexadecimal example (Figure 4-15) defines the User Data field of a CTRL_RSP_LE frame that indicates a T1 (Delay Before Transmission) value of 11 MAC cycles, an SQP (Signal Quality Parameter) value of 15, and support for a CLNP interface with the CT1 (Multicast Packet Period) parameter set to 10 packets:

82 80 00 14 01 0F 38 38 38 35 3A 31 39 39 33 3A

FI GI GL PID PL PV

Public Group Parameter Set ID

4D 6F 64 65 33 09 01 0B F0 00 11 00 01 56 02 01

PV (cont’) PID PL PV GI GL PID PL PV PID PL

Parameter set ID (cont’d) T1 Timer Private group Parameter set ‘V’ SQP

- 27 –

0F 00 01 57 02 01 01 01 42 02 00 0A

PV PID PL PV PID PL NT NL SID SL SV

Parameter set ‘V’ Network Initialization

LEGEND: FI - Format IdentifierGI - Group IdentifierGL - Group Length (in bytes)PID - Parameter IdentifierPL - Parameter LengthPV - Parameter ValueNT - Network TypeNL - Network LengthSID - Subparameter IdentifierSL - Subparameter LengthSV - Subparameter Value

Figure 4-15. User data field of a CTRL_RSP_LE frame (Hexadecimal example)

4.15 FLEXIBLE SUBNETWORK INTERFACE

4.15.2The system currently has defined interfaces for communicating ISO 8208 packets, for communicating CLNP packets, and for raw data. The ISO 8208 interface supports segmentation of packets as needed, while the CLNP interface currently does not. The raw interface is primarily intended for MOPS and certification testing of hardware, but could also be used if a State desired a non-networked ATN subnetwork service. The raw interface also supports FIS-B and ACARS messaging via extended ISO 9577 signaling.

Note. — If segmentation is needed in the CLNP interface to support ATN communications, either the Compression Type field can be expanded to indicate the inclusion of a Version field, which in turn would allow for a version to include Segmentation, or a new Network Type for a modified CLNP interface that includes segmentation could be added.

4.16.3 Raw subnetwork compression

The raw subnetwork interface does not currently support compression. The raw interface makes use of ISO 9577 signaling to indicate the contents of the user data.

4.18.2For data-capable configurations (e.g. 2V2D, 3V1D, 2V1D, 1V3D, 1V2D and 3T), the ground station will send reservation messages to every Local ID in ascending order of Aircraft ID. Aircraft users will respond to reservations addressed to them with CTRL_CMD_LE frames containing the Expedited Recovery XID and the Network Initialization XID.

4.18.3While the CTRL frame process is ongoing, the ground station will poll every Local ID in descending order of Aircraft ID. Aircraft users will respond to polls addressed to them by sending a Net Entry Request message in the next MAC cycle. To speed up the recovery process, the list of Local IDs to whom reservations are sent is pruned by using information gleaned from the poll response procedure. The polling process can cease wheneverOnce the Local ID being polled is numerically less than the Local ID currently receiving a reservation, the ground station is only looking for enhanced-voice-only aircraft radios. Note that random access (RA) downlink M channel messages such as Reservation Requests are not allowed during the recovery process. Upon completion of the CTRL frame process, the ground station will broadcast a CTRL frame with the Connection Check XID indicating all of the aircraft it has

- 28 –

recovered. If an aircraft determines that the information listed is incorrect, it will send a Leaving Net message and then attempt to re-enter the net. If an aircraft does not find itself on the list when it believes it should be, it will attempt the normal Net Entry procedure when the ground station returns to the Normal uplink message.

4.18.4For voice-only configurations (4V, 3V, 3S, and 2S1X), the CTRL frame process is not applicable and the recovery process consists of just the polling process. If the voice-only configuration is not supporting discrete-addressing, there is no need to use the expedited recovery procedures.

4.18.5If an aircraft radio does not receive a poll or a reservation during the recovery process or it is not contained within the Connection Check XID broadcast out at the end of the recovery procedure, it will assume that its connection is no longer valid, and it will attempt net entry when normal operation of the ground radio resumes.

4.19 GROUND STATION FAILURE

If the ground station fails, all of the discrete addressing information and data scheduling state information could be lost. Upon restoration of the ground station, voice service should be restored to Timing State 1 as soon as the beacon information transmits twice. To restore data transmission quickly, the system will undergo an expedited recovery procedure, per Section 4.18. The ground station will uplink the Recovery M burst granting data access to two potential aircraft per MAC cycle, with one V/D (data) slot assigned to each aircraft. The reservation grants will be conducted sequentially from local ID 1 to 60, as the ground does not yet know which local identifiers are active. This will take 30 MAC cycles to complete and will allow the ground station to know which aircraft are in the net. The aircraft will send a CTRL_RSP frame including the Expedited Recovery XID and Network Initialization XID. If the ground station does not receive any response on an assigned slot, it will assume no active aircraft has been assigned that local ID. After all 60 local IDs have been stepped through, the ground station will send a CTRL_RSP_LPM frame containing the Connection Check XID that lists the local IDs of all the aircraft from which it received a response. It will then return to sending normal uplink beacons and begin granting data slots as requested to re-establish the network connections. If an aircraft does not find itself in the Connection Check XID, it should attempt the normal link establishment procedure to restore the connection.

Part I – Implementation aspectsChapter 5 – MAC-DLS operations

5.4.2For each incoming DLS frame, the DLM performs error detection using the frame check sequence (FCS). For a discretely addressed frame sub-group with one or more corrupted FCS, the entire frame sub-group will be rejected by the DLS. Only when all DLS frames in a discretely addressed frame group pass the FCS, will the DLE generate an acknowledgment frame for transmission to the source DLE. Upon receiving the acknowledgment frame, the source DLS will immediately inform the MAC sub-layer to cancel its T1 timer and cancel any action or preparation for retransmission. A new frame or frame sub-group from a specific DLE may be sent to the MAC sub-layer through the DLM whenever the MAC sub-layer is ready for next media access event, and there is no T1 timer active associated with the DLE. Upon expiration of the T1 timer, the MAC sub-layer declares time-out for the transmitted frame group and informs the DLM to provide a new frame group. This new frame group will include either the outstanding frame sub-group or a new frame sub-group with higher priority than that of the outstanding frame sub-group to initiate the next transmissionminus any unacknowledged frames from previous transmissions and possibly some new unacknowledged frames. Interlayer protocol primitives may be used by the DLS and MAC to correlate events and to convey the priority level and size of each DLS frame (or frame group) sent between the two sub-layers.

5.7.1.9The MAC_time_out.IND primitive is used by the MAC sub-layer to inform the DLS sub-layer that the T1 timer associated with a frame group transmitted earlier has expired indicating that an ACK message has not been received within the allowable time. Upon receiving the MAC_time_out.IND, the DLE will terminate its WAIT state for the acknowledgment frame and initiate a MAC-data.REQ protocol primitive which carries the highest priorityoutstanding frame sub-group (a new frame sub-group or the outstanding frame sub-group) to the DLM immediately. The DLM will generate a new frame group and send the new frame group to the MAC sub-layer.

5.7.2 Operations of a downlink MAC sub-layermedia access event

The following sequence of operations describes a normal downlink MAC media access event:

[TEXT SKIPPED]

D14: DLS frame group retransmission

If the T1 timer expires without receiving the MAC_cancel_t1.REQ primitive, the MAC will declare a time-out for the outstanding frame group by sending a MAC_time_out.IND protocol primitive to the DLE through the DLM. Upon receiving the MAC_time_out.IND, the DLE will terminate its WAIT state for the acknowledgment frame and initiate a MAC_data.REQ protocol primitive which carries the highest priorityoutstanding frame sub-group (a new frame sub-group or the outstanding frame sub-group) to the DLM immediately. The DLM will generate a new frame group and send the new frame group to the MAC sub-layer. Upon receiving the new frame group the aircraft MAC sub-layer will prepare a reservation request for the received frame group.

5.7.3 Operations of an uplink MAC sub-layer media access event

The following sequence of operations describes a normal uplink MAC media access event: [TEXT SKIPPED]

U11: DLS frame group retransmission

If the T1 timer expires prior to receiving the MAC_cancel_t1.REQ primitive, the ground station MAC will declare a time-out for the outstanding frame group by sending a MAC_time_out.IND protocol primitive to the source DLE through the DLM. Upon receiving the MAC_time_out.IND, the DLE will terminate its WAIT state for the acknowledgment frame and initiate a MAC_data.REQ protocol primitive which carries the highest priorityoutstanding frame sub-group (a new frame sub-group or the outstanding frame sub-group) to the DLM immediately. The DLM will generate a new frame group and send it to the MAC sub-layer. The ground station MAC will prepare itself for the transmission of the frame group.

Part I – Implementation aspectsChapter 8 – Time maintenance

8.3.3.1In order to maintain time, the aircraft radios must keep track of three separate counters:

a) Coast Time Counter 1 (CTC1) is the number of MAC cycles since the last successful reception of an uplink M channel beacon. Thus, whenever a beacon is missed CTC1 is incremented by 1. Otherwise, CTC1 is reset to 0. A successful beacon reception consists of the reception of the synchronization sequence and the Beacon 2 word. The Slot NumberBeacon ID and GSC fields must correspond to the aircraft radio’s net. In order to qualify as a valid beacon signal the beacon must also contain the same information in Beacon 2 and have the same timing (within ±1 symbol period) as the previous beacon signal.

b) Coast Time Counter 2 (CTC2) is the number of MAC cycles since the last time the aircraft radio accepted new timing information. This information could come from a beacon reception, but it could also come from the reception of an Alternate Timing Signal (ALTS). ALTSs are Poll Responses from aircraft radios (except in the 3T configuration) and beacons in slots other than the one controlling the aircraft radio’s User Group. Different User Groups sharing a common channel may be controlled by geographically separated ground stations. However, the separate ground stations will have the same GSC. Aircraft radios will accept the timing information of an ALTS only if it obeys certain timing constraints described below in rule b) of 8.3.4.1. If a beacon is not received and if the timing is not updated by an ALTS then CTC2 is incremented by 1. Otherwise, CTC2 is reset to zero.

c) Coast Time Counter 3 (CTC3) is the number of MAC cycles since the last reception of a beacon or any Poll Response whether or not it was used for updating time. If none of the above signal types is received during a MAC cycle, CTC3 is incremented by 1. Otherwise, CTC3 is reset to 0.


- 37 -

8.3.3.4In order to facilitate the efficient use of aircraft radios, it may be possible advantageous to indicate to the users (by a means to be determined) their current timing states.

8.3.4 Transition rules

8.3.4.1This section will describe the rules whereby the Coast Time Counters (CTCs) are used to control the transitions from one timing state to another and the timing updates using the ALTSs. In order to derive these rules certain assumptions about the system parameters were made:

a) The stability of the aircraft clock is the same as the aircraft frequency stability, i.e. 5 ppm.

b) The ground stations are normally locked to an external time source. Under very rare circumstances the ground radio may have a timing accuracy equivalent to its frequency stability, i.e. 2 ppm.

c) An additional source of time error is the movement of the aircraft radio with respect to the ground station. For a Mach 1 aircraft this is 1 ppm.

d) One type of ALTS is the polling response. In the worst case of a fully loaded 4V configuration, the longest possible polling cycle is 120 MAC cycles (plus some allowance for net entrants).

8.3.4.3Given these assumptions, the following rules for aircraft radios which have already entered a net (and are not in the 3T configuration) can be generated:

- 38 -

a) When CTC1 t (default =50), a radio will change from TS1 to TS2.

b) While in TS2, a radio will accept time from an ALTS if, and only if,

Tr < Ta + 0.04 CTC2 Ts.

Here, Ta is the expected time of an ALTS as determined by the aircraft radio’s clock; and T r is the actual time of arrival of the ALTS with respect to the aircraft radio’s clock. Ts is a symbol period. If CTC2 is greater than 800, all ALTS time updates will be accepted. If the time is updated, CTC2 is reset to 0.

c) When CTC3 f (default = 250), a radio will change to TS3.

d) While in TS3, a radio will accept time from any ALTS and enter TS2.

e) Any radio in TS2 or TS3 will immediately revert to TS1 upon reception of a valid beacon signal. Time will be updated to agree with that of the beacon.

Thus, CTC1 and CTC3 control a radio’s timing state whereas the role of CTC2 is to control the timing update mechanism while a radio is in TS2. A “state diagram” based on these rules is shown in Figure 8-2.

8.3.4.3.2Rule b) is designed to allow an aircraft radio to accept time preferentially from sources with the least excess path delay. Thus, a radio will accept time from an ALTS only if it appears “earlier” than the time predicted by its own internal clock. The factor with CTC2 is included to allow for the case where an aircraft radio’s clock is fast with respect to the correct time. The net result is that, when there are a number of ALTSs available, an aircraft radio will tend to have its time controlled in such a way as to minimize its excess path delay. This will greatly reduce the chance that a radio in TS2 will ever violate its time slot boundaries.

8.3.5 Special rules for net initialization

8.3.5.1Radios which have not yet entered a net have some special considerations when uplink beacons are not available. Such radios can be considered to be in a special timing state, called TS0, where transmission is not allowed. Radios in this state have no information as to the local configuration of the system, and they initially have no information as to why beacon signals are not available. The ground station could be operating but out of line of sight, or the ground station could be inoperable. To deal with this uncertainty, the following rule applies. The aircraft radio will not transmit until one of the following happens:

a) The radio receives a valid beacon signal. In this case the word “valid” signifies that the information in Beacon 2 is the same and the timing is the same (within ±1 symbol period) as the last successful beacon reception. In this case the radio will enter TS1 and proceed to net entry if desired.

b) The radio receives an ALTS. The ALTS will contain enough information (see Chapter 3, 3.5) to tell the aircraft radio whether the local configuration is a 3-slot or 4-slot configuration. With this information the radio can enter TS2 and determine the correct time for transmitting truncated voice messages.

c) The radio’s CTC3 counter reaches the value f. In this case the radio enters TS3. In order to shorten the time it takes to reach TS3 under these circumstances, the CTCn are initially set to max(f-50,t) at the onset of link establishment. Using the default values of f=250 and t=50 gives 200 as the starting

- 39 -

value of CTCn. In this case an aircraft radio will enter TS3 in 12 seconds when no beacon is present.

8.3.7.2Note that in 3T there are no valid ALTSs since there are no alternative uplink M channel beacons and the Poll Responses are identified as not being suitable for timing (see 2.4). However, the Poll Responses do serve to reset CTC3 to zero whenever they are received. A potential net entrant entering into a 3T area will never have the opportunity to enter via TS2 since no ALTSs are available. It will either enter normally in TS1 or go directly to TS3 after waiting up to one minute.

8.3.9 Scenarios

To clarify how the proposed timing rules will work, three possible “lost timing” scenarios are considered.

a) Single lost radio. In this scenario a single aircraft radio in a user group loses connectivity with the ground station due to some type of adverse propagation condition. All (or most) other users remain connected. In this case the lost radio will coast for 12 seconds and then enter TS2, the truncated voice mode. While in this mode, it can maintain voice communication with some of the other aircraft radios. It will also receive poll responses from some of the radios. The radio will update its timing based on the “closest” relay radios available. (Closest connotes, in this case, radios which are most nearly in line with the ground radio and the lost radio. These will have the smallest excess path delay.) This will, with very high probability, keep the truncated voice transmissions within their proper time slot boundaries. Also, the quasiperiodic reception of ALTSs will keep the lost radio from ever entering TS3. In most cases, the lost radio will eventually re-establish its connection with the ground. Reception of the ground beacon will reset all CTCs to zero and send the radio into TS1.

b) Massive Ground Failure. In this scenario the ground station and all possible back-ups fail simultaneously. When this happens, the CTC1’s of all aircraft radios will begin to increment once per cycle. After 12 seconds all the radios will enter TS2. Also, all CTC3 counters will continue to increment. After about 60 seconds, all the radios will enter TS3 and all user groups will coalesce into one large user group. At no time is air-to-air voice communication prevented. Also, if and when the ground stations are operable and begin transmitting beacons, the aircraft radios will all return to TS1.

c) Oceanic. In this case, which might apply over the ocean or in any area with no ground infrastructure, there are never any beacon signals. The only correct timing state is TS3. In this case any entering radio will wait 12 seconds for its CTC3 to reach 250, and it will then remain in TS3 indefinitely.

8.3.10 Dummy poll responses

8.3.10.1The overall system timing architecture relies on air-to-air time transfer provided by quasiperiodic Poll Responses. In particular, the CTC3 thresholds for entering TS3 are determined based on assumptions concerning typical polling cycle times. However, there exist certain configurations in which polling may not be implemented. These configurations are the voice-only configurations: 4V, 3V, 3S and 2S1X. If a ground radio in one of these configurations does not support polling, it will transmit the value 61 or 62 in the Aircraft ID (Poll) field of its uplink M channel message. This informs the aircraft radios that discrete addressing is not supported.

8.3.10.3Immediately after entering TS1, after determining that the uplink Beacon contains Aircraft ID = 61 or 62, divide all further MAC cycles into groups of 25. Within each group, choose one MAC cycle at

- 40 -

random in which to transmit a dummy Poll Response. Continue to transmit dummy Poll Responses for as long as the aircraft radio remains in TS1.

Table 8-1. Source filtering squelch window matrix for 4-slot configurations

Aircraft Receiver State Squelch Window (Note 1)

Ground Receiver State Squelch

Window (Note 1)

StateMessage Type

TS1(Note 2)

TS2(Note 2, Note 3)

TS3(Note 5)

N/ATS1(Note 2)

Uplink voice -1 to +1 -33 to +1 Wide Open N/A

Downlink voice: TS1 -2 to +4(n+1) -34 to +4(n+1) Wide Open -1 to +(4n+3)


Downlink voice: TS3Wide Open

(Note 4)Wide Open

(Note 4) Wide OpenWide Open

(Note 4)

Note 1.— Each squelch window is defined relative to the nominal burst time in symbols.Note 2.— n = Squelch Window parameter.Note 3.— If n is unknown, then the default value is 6.Note 4.— Receive any such message with “correct” Group ID if not already receiving a voice message.Note 5.— In this column, “Wide Open” implies that an aircraft radio will receive any voice message, without regard to Group ID.

- 41 -

Part I – Implementation aspectsChapter 9 – Miscellaneous items

9.1 RANDOM NUMBER GENERATOR

Note.— See Appendix C of the Technical Provisions of this Manual.9.1.1Certain functions of the radio system related to RA require the generation of a “random” number. For instance,

in all configurations the radio normally must choose which slot to use for a Net Entry Request. To ensure that different radios generate different numbers, the algorithm for generating random numbers is based on the unique 24-bit ICAO address.

9.1.2The algorithm generates a new 12-bit number at each access as follows. First, the 24-bit address is broken into two 12-bit pieces so that

[delete formula]

These are used to generate two sequences of numbers using the recursion relations:

with [delete formulas] These are then combined to give a single random number:

This ensures that two radios will not choose the same pseudorandom sequence.

9.1.3This algorithm produces 224 unique sequences, each of which is nearly uniform in distribution. These sequences can be used to derive sequences of numbers drawn from a smaller alphabet. Suppose, for example, that a sequence of numbers between 0 and 8 (inclusive) must be generated for random access in configuration 2V2D. Such a number can be derived from K(n) by invoking the formula:

where [delete formula] means the integer part of the argument.


[delete formulas]

[delete formula]

[delete formula]

- 44 –

9.3 GROUND STATION USER LIST

9.3.1In order to manage the interface between the VDL Mode 3 radio network and the ground network, the ground station maintains a list of current active users on each net. This list consists of a record of the unique relationship between an aircraft user’s 24-bit ICAO address, 2-bit Radio ID and its 8-bit Local ID. Since the radio access is controlled by the Local ID and network access is controlled by the ICAO address, a translation must sometimes be made at the interface. Since an airframe may have multiple radios, the Radio ID is used to determine the specific Local ID assigned to the unit from the ICAO address, which is only unique to the airframe. For example, during an aircraft voice transmission, the identity of the speaker can be sent to the ground controller by translating the Local ID in the V/D header into the corresponding ICAO address. The ICAO address can, in turn, be translated into the aircraft’s call sign if that information is available to the ground network.

9.3.2Whenever a user is added to or deleted from the user list, an appropriate message is sent to a higher-level protocol in the ground system. A user is added to the list during the net entry procedure whenever a ground radio successfully receives the first Poll Response message during net entry. A user is deleted from the list whenever the ground station receives a Leaving Net message or generates a Terminate Net message. A user can also be deleted from the list if it fails to respond to NL2 (default = 3) poll commands. This type of purging can either be automatic or manual, depending on the ground system implementation.

9.3.3Note that the user list can also be used to avoid ambiguities that could arise if a user who is already in a net should try to re-enter it. (This could happen if a net entrant failed to receive a Reservation message immediately following the first Poll Response.) In that case, the “entrant” will be reissued thea new Local ID and the Radio ID will be incrementedthat is already in the list. If it is a duplicative entry in the net, the old Local ID association will time out after NL2 polls are not answered.

9.4.1 Link establishment

9.4.1.1The link establishment procedure consists of several steps. In the net initialization phase, the entrant must receive an uplink beacon twice in a row with the identicalconsistent (correct) information in the Beacon 2 word before it can begin the net entry phase. Note that this word contains a GSC, which should virtually eliminate the possibility of entering a nearby net using the same frequency as the correct net.

9.4.3 Poll responses

A ground station may fail to correctly receive poll responses when they are commanded. This may be due to the fact that the aircraft radio has left the net. After NL2 (default = three) consecutive missed poll responses, the ground radio will notify a higher-level protocol and the aircraft radio will be a candidate for purging.

9.4.4.2There is a very small, but finite, chance that a Leaving Net message will be received erroneously. The nine extra bits in the message are set to 0 to help identify when the message might be corrupted. It may happen that the received message has an incorrect Local ID that corresponds to another user on the net. If this happens, the user will incorrectly be dropped from the net and deleted from the Ground Station User List. When it is dropped from the net, the user will stop receiving poll commands; however, it can re-enter the net using the procedures described in 9.4.5.

- 45 –

9.5.2The first mechanism relies on the fact that the ground station has a validity window for downlink M channel messages which is the same size as the squelch window. Thus, if an aircraft entrant’s range is substantially beyond the sector size, its Net Entry Request will not be valid, and the ground will not respond. Thus, the aircraft radio will not be able to enter the net and will not be allowed to transmit Reservation Requests, V/D (data) messages, etc. Nevertheless, the aircraft radio could still attempt to transmit V/D (voice) messages using the dummy Local Aircraft ID (= 61 or 62).

9.6 BACK-UP GROUND SITES

9.6.1For a variety of reasons, it may happen that the ground site that controls a particular user group will change abruptly. This might occur, for example, if there is a failure at a primary site which causes the ground infrastructure to switch to a back-up radio site. The validity window is now large enough to allow the site to change within a reasonable distance and timing to be maintained. If the back-up site is sufficiently far from the primary site, the new beacon signals may appear (at some aircraft receivers) to be outside the ±1-K(n+1)+1 to +1 symbol period validity window. Thus, the first such beacon will be declared invalid, and CTC1 and CTC2 will be incremented. (CTC3 will reset to 0.) However, if the next received beacon has the same time offset, it will be declared valid since the information (in this case the timing information) is the same.

9.6.2Thus, whenever an aircraft radio receives two successive beacon signals with timing information which is substantially different from its previous information, the timing of the aircraft radio should be changed. This will allow the aircraft radio to quickly adapt to the new geometry with a minimal likelihood of false time adjustment. Note that by this same mechanism other information contained in the beacon, such as the Squelch Window parameter or the Configuration, can be changed.

9.6.3Recall to support diversity site group operations, the aircraft radio will utilize the validity window to determine whether timing should be updated, but will process the instructions from M uplink messages that have the appropriate GSC.

Part I – Implementation aspectsChapter 12 – Miscellaneous items

12.1 INTRODUCTION

12.1.1Care has been taken in this manual to define each burst component as exactly as possible. Nevertheless, there may be some ambiguity as to the order in which the different pieces are assembled and the order of the bits (or symbols) within the pieces. To help remove such ambiguity, an example of a particular downlink V/D (data) message is provided. Such a message contains all types of burst components including a synchronization sequence, a Golay code word and a Reed Solomon (RS) code word.

12.1.2It is assumed for the sake of this exercise that a data frame comprising exactly 62 bytes (octets) of information is being sent. This means that the entire message consists of one, and only one, burst. It is also assumed that the aircraft radio making the transmission is connected to a ground site whose Ground Station Code (GSC) is 010.

12.1.3Throughout this chapter, numerous examples of bit or symbol sequences are presented in tabular form. These are always read from left to right, and then from top to bottom (exactly as one would read this text). Ultimately, the information is transmitted in this left-to-right fashion. For all MAC sub-layer and physical layer processing, the most significant bit (MSB) of any field is placed to the left and the least significant bit (LSB) is placed to the right.

12.1.4The different components of a burst will be treated separately in the following paragraphs. Then the pieces will be put together, and the further processing needed to create the final Differentially encoded 8 Phase Shift Key (D8PSK) transmission will be described.

12.2 POWER STABILIZATION

It is assumed that every burst begins with a sequence of three symbols with information 000. The purpose of this sequence is to provide a more or less constant signal to allow a receiver’s Automatic Gain Control (AGC) to stabilize prior to receiving the first symbol of the synchronization sequence. This corresponds to the 9-bit sequence shown in Figure 12-1.

0 0 0 0 0 0 0 0 0

Figure 12-1. Power stabilization sequence

12.3 SYNCHRONIZATION SEQUENCE

The synchronization sequence for all V/D (data) bursts is S2 (as described in Chapter 3, 3.2), which is shown in Figure 12-2.

Figure 12-2. Synchronization Sequence

12.4 HEADER

As described in Chapter 2, 2.2, the V/D (data) header has the following format:

{(Message ID:3)(Ground Station Code:3)(Segment Number:4)(Spare:1)(EOM:1)}

(17 pages)ACP-WGM-WP-09.doc

- 49 -

0 0 0 1 1 1 0 1 1 0 1 0 0 0 0 1 0 0 0 0 1 0

1 0 0 1 0 1 0 1 1 1 1 0 0 0 1 1 1 0 1 0 1 1

Figure 12-2. Synchronization sequence

In this case the fields have the following values:

Message ID = 101 Downlink dataGSC = 010 By assumptionSegment Number = 0001 Data frame is one burst longSpare = 0 All spare bits are set to 0EOM = 1 Data frame is one burst long

Thus, the header for this particular burst is:

1 0 1 0 1 0 0 0 0 1 0 1

This header is protected by a (24,12) Golay code which can be implemented by multiplying the previous 12-element vector by a particular 12 x 24 matrix to yield Figure 12-3. Note that in this case the Golay parity bits are affixed ahead of the original header bits.

0 1 1 0 0 0 0 0 0 1 0 0 1 0 1 0 1 0 0 0 0 1 0 1

Figure 12-3. Golay coded header sequence

12.5 DATA AND RS ERROR CODINGFRAME

For this example it is assumed that the data (as interpreted by the RS encoder)frame being sent from air to ground ahs the following content consist of the following 62 bytes of information (in hexadecimal notation):

AA AA A581 00 1A82 F0 0016 00 0157 02 0301 03 5201 05 5302 00 B454 02 0014 03 0004 00

00 01 02

03 04 05

06 07 08

09 0A 0B

0C 0D 0E

Rob Morgenstern, 06/14/05,

Editors Note: Delete the empty rows. Only left to illustrate deletion

- 50 -0F 10 11

12 13 14

15 16 17

18 19 1A

1B 1C 1D

1E 1F 20

21 22 23

24 25 26

27 28 29

2A 2B 2C

2D 2E 2F

30 31 32

33 34 35

36 37 38

39 3A 3B

3C 3D

The first two rows contain the frame header information, and the remainder is 26 bytes of User Data. At this point the ICAO address (first row) has not been combined with the Frame Check sequence. The interpretation of the header information is as follows:

ICAO Address 101010101010101010100101 Address Type 001 = Frame transmitted by specified aircraft Toggle Bit (T) 0 More Frame (MF) 0 Frame Type 001 = CTRL_CMD (Control Command) Ground Subnet Address 000000 Data Length 0000011010 = 26 bytes

(Note that for the Address Type field the MSB is the third bit instead of the first one.)

The 26 bytes of user data have the following significance:

82 General Identifier (ISO 8885)F0 Private00.16 Format Length = 22 bytes00.01.57Private Parameter Set ID = W (VDL Mode 3 Private)02.03 Network Initialization Parameter (3 subnetwork types) 01.03 Compressed 8473 (3 subparameters) 52.01.05 CT1 = 5 53.02.00.B4 CT2 = 180 54.02.00.14 CT3 = 20 03.00 Uncompressed 8473/9542 (no subparameters) 04.00 Raw (no subparameters)

For convenience, it has been assumed that the 62 bytes are just the ascending sequence of numbers from 0 to 61 (in decimal notation); an actual data frame would not be likely to have such a simple form. Note, also, that in this example the data frame consists of exactly 62 bytes. If there had been less data, then it would have been padded with zeros until the total length was 62 bytes. These padded zeros are always appended to the end (i.e. to the right) of the data and transmitted along with the rest of the message. All V/D (data) bursts are the same length.

- 51 -The RS encoding process generates 10 bytes of parity information. For the input above, the parity bytes are (0E, DC, 13, 63, E7, 4A, 6D, 3E, 1D, A2). These are affixed to the end of the data. Thus, the entire 72-byte RS word is:

00 01 02

03 04 05

06 07 08

09 0A 0B

0C 0D 0E

0F 10 11

12 13 14

15 16 17

18 19 1A

1B 1C 1D

1E 1F 20

21 22 23

24 25 26

27 28 29

2A 2B 2C

2D 2E 2F

30 31 32

33 34 35

36 37 38

39 3A 3B

3C 3D 0E

DC 13 63

E7 4A 6D

3E 1D A2

This is expressed in terms of bits in Figure 12-4.

0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 1 0 0 0 0 0 0 1 00 0 0 0 0 0 1 1 0 0 0 0 0 1 0 0 0 0 0 0 0 1 0 10 0 0 0 0 1 1 0 0 0 0 0 0 1 1 1 0 0 0 0 1 0 0 00 0 0 0 1 0 0 1 0 0 0 0 1 0 1 0 0 0 0 0 1 0 1 10 0 0 0 1 1 0 0 0 0 0 0 1 1 0 1 0 0 0 0 1 1 1 00 0 0 0 1 1 1 1 0 0 0 1 0 0 0 0 0 0 0 1 0 0 0 10 0 0 1 0 0 1 0 0 0 0 0 0 0 1 1 0 0 0 1 0 1 0 00 0 0 1 0 1 0 1 0 0 0 1 0 1 1 0 0 0 0 1 0 1 1 10 0 0 1 1 0 0 0 0 0 0 1 1 0 0 1 0 0 0 1 1 0 1 00 0 0 1 1 0 1 1 0 0 0 1 1 1 0 0 0 0 0 1 1 1 0 10 0 0 1 1 1 1 0 0 0 0 1 1 1 1 1 0 0 1 0 0 0 0 00 0 1 0 0 0 0 1 0 0 1 0 0 0 1 0 0 0 1 0 0 0 1 1

- 52 -0 0 1 0 0 1 0 0 0 0 1 0 0 1 0 1 0 0 1 0 0 1 1 00 0 1 0 0 1 1 1 0 0 1 0 1 0 0 0 0 0 1 0 1 0 0 10 0 1 0 1 0 1 0 0 0 1 0 1 0 1 1 0 0 1 0 1 1 0 00 0 1 0 1 1 0 1 0 0 1 0 1 1 1 0 0 0 1 0 1 1 1 10 0 1 1 0 0 0 0 0 0 1 1 0 0 0 1 0 0 1 1 0 0 1 00 0 1 1 0 0 1 1 0 0 1 1 0 1 0 0 0 0 1 1 0 1 0 10 0 1 1 0 1 1 0 0 0 1 1 0 1 1 1 0 0 1 1 1 0 0 00 0 1 1 1 0 0 1 0 0 1 1 1 0 1 0 0 0 1 1 1 0 1 10 0 1 1 1 1 0 0 0 0 1 1 1 1 0 1 0 0 0 0 1 1 1 01 1 0 1 1 1 0 0 0 0 0 1 0 0 1 1 0 1 1 0 0 0 1 11 1 1 0 0 1 1 1 0 1 0 0 1 0 1 0 0 1 1 0 1 1 0 10 0 1 1 1 1 1 0 0 0 0 1 1 1 0 1 1 0 1 0 0 0 1 0

12.6 BIT STUFFING AND FRAME CHECK SEQUENCE

12.6.1 The next stages in the construction of a message are bit stuffing and the addition of the Frame Check Sequence (FCS). Bit stuffing is a simple process of adding enough zeros at the end of the message so that the message contains exactly 62 bytes. The FCS is a 24-bit sequence generated using the CRC-24Q polynomial, computed over the entire frame. The polynomial used is as follows:

x24 + x23 + x18 + x17 + x14 + x11 + x10 + x7 + x6 + x5 + x4 + x3 + x+1

For the particular bit sequence of the previous section the FCS is given by

000111011110011000101010 = 1DE62A

The FCS is overlaid on the 24-bit ICAO Address with an exclusive-OR operation.

12.6.2 The result of bit stuffing and the FCS overlay is as follows:

B7 C4 8F81 00 1A82 F0 0016 00 0157 02 0301 03 5201 05 5302 00 B454 02 0014 03 0004 00 0000 00 0000 00 0000 00 0000 00 0000 00 0000 00 0000 00 0000 00 0000 00 00

- 53 -00 00

12.7 BIT SWAPPING AND RS ENCODING

12.7.1 The next phase of the message generation process is bit swapping. This step is necessary because the bottom layers of the protocol stack use an interpretation of MSB and LSB that is reversed from that used in the higher layers. Thus, the previous 62-byte block becomes the following:

ED 32 F181 00 5841 0F 0068 00 80EA 40 C080 C0 4A80 A0 CA40 00 2D2A 40 0028 C0 0020 00 0000 00 0000 00 0000 00 0000 00 0000 00 0000 00 0000 00 0000 00 0000 00 0000 00

12.7.2 The RS encoding process generates 10 bytes of parity information. For the input above, the parity bytes are (6A, 46, B2, 9F, 1C, EE, B3, 25, EB, F4). These are affixed to the end of the data. Thus, the entire 72-byte RS word is as follows:

ED 32 F181 00 5841 0F 0068 00 80EA 40 C080 C0 4A80 A0 CA40 00 2D2A 40 0028 C0 0020 00 0000 00 0000 00 0000 00 0000 00 0000 00 0000 00 0000 00 00

- 54 -00 00 0000 00 0000 00 6A46 B2 9F1C EE B325 EB F4

This is expressed in terms of bits in Figure 12-4.

1 1 1 0 1 1 0 1 0 0 1 1 0 0 1 0 1 1 1 1 0 0 0 11 0 0 0 0 0 0 1 0 0 0 0 0 0 0 0 0 1 0 1 1 0 0 00 1 0 0 0 0 0 1 0 0 0 0 1 1 1 1 0 0 0 0 0 0 0 00 1 1 0 1 0 0 0 0 0 0 0 0 0 0 0 1 0 0 0 0 0 0 01 1 1 0 1 0 1 0 0 1 0 0 0 0 0 0 1 1 0 0 0 0 0 01 0 0 0 0 0 0 0 1 1 0 0 0 0 0 0 0 1 0 0 1 0 1 01 0 0 0 0 0 0 0 1 0 1 0 0 0 0 0 1 1 0 0 1 0 1 00 1 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 1 0 1 1 0 10 0 1 0 1 0 1 0 0 1 0 0 0 0 0 0 0 0 0 0 0 0 0 00 0 1 0 1 0 0 0 1 1 0 0 0 0 0 0 0 0 0 0 0 0 0 00 0 1 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 00 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 00 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 00 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 00 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 00 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 00 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 00 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 00 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 00 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 00 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 1 1 0 1 0 1 00 1 0 0 0 1 1 0 1 0 1 1 0 0 1 0 1 0 0 1 1 1 1 10 0 0 1 1 1 0 0 1 1 1 0 1 1 1 0 1 0 1 1 0 0 1 10 0 1 0 0 1 0 1 1 1 1 0 1 0 1 1 1 1 1 1 0 1 0 0

Figure 12-4. Reed-Solomon Coded Data Sequence

12.7.3 It should be emphasized that the purpose of the RS encoding and decoding process is to ensure that the 496 bits of data carried by the burst are delivered as reliably as possible from the Data Link Service (DLS) sublayer of the transmitter to the DLS sublayer of the receiver. The fact that the RS coder and RS decoder interpret these bits as a sequence of 62 bytes with the MSB on the left does not have any relevance at all to how these bits may be interpreted by the upper layers of the transmitting and receiving protocol stacks.

Figure 12-4. Reed-Solomon coded data sequence

It should be emphasized that the purpose of the RS encoding and decoding process is to ensure that the 496 bits of data carried by the burst are delivered as reliably as possible from the Data Link Service (DLS) sub-layer of the transmitter to the DLS sub-layer of the receiver. The fact that the RS coder and RS decoder interpret these bits as a sequence of 62

- 55 -bytes with the MSB on the left does not have any relevance at all to how these bits may be interpreted by the upper layers of the transmitting and receiving protocol stacks. The bits are simply delivered in the order in which they are received.

12.68 BIT SCRAMBLING

12.8.1 The payload of every VDL Mode 3 burst is subject to bit scrambling. Bit scrambling consists of bit-by-bit modulo 2 addition with a particular sequence of bits. The sequence is part of an m-sequence defined by the characteristic polynomial

Equivalently, the sequence is defined by the recursion relation

with a seed (s(1),s(2), s(15)) equal to (1,0,0,1,1,0,1,0,1,0,0,1,0,1,1). The scrambling bits are then given by s(16) through s(615) as shown in Figure 12-5. This 600-bit sequence is fixed, and every burst is scrambled using this same sequence. The V/D (data) and normal V/D (voice) bursts are the longest ones, and they use all 600 bits in the sequence. Other types of VDL Mode 3 bursts are shorter, and they use only as many of the scrambler bits as are needed, starting at the beginning of the sequence.

0 0 0 1 0 0 1 1 0 0 0 1 1 0 1 1 1 1 0 0 0 1 0 00 0 1 0 0 1 0 1 0 0 0 0 1 1 1 1 1 0 0 0 1 1 0 00 0 0 1 0 1 0 1 1 1 1 0 1 1 1 1 1 1 0 0 1 1 0 10 1 1 0 1 0 1 0 1 1 1 0 1 1 0 0 1 0 0 1 1 0 0 10 1 1 0 1 1 1 0 0 0 1 0 0 0 1 1 0 1 1 0 1 0 0 00 1 1 1 1 0 1 1 0 1 1 0 0 0 0 0 1 0 1 0 0 1 0 01 0 0 0 0 0 0 1 1 0 0 0 1 1 1 0 0 0 0 0 0 0 1 00 0 0 1 0 1 1 1 1 1 1 1 1 0 0 0 0 0 1 1 0 1 0 10 1 0 1 1 1 1 1 1 0 1 1 0 0 1 1 0 0 1 0 1 0 1 00 1 0 0 0 1 0 0 0 1 1 0 0 1 1 1 0 0 0 0 1 1 1 10 1 1 1 0 1 0 0 0 0 0 1 0 1 0 0 1 0 1 1 0 0 0 00 0 1 1 0 0 0 1 1 0 1 1 1 1 1 1 1 0 1 1 1 1 0 11 0 1 0 1 0 1 0 0 1 0 1 0 0 1 0 0 1 1 0 0 1 1 10 0 1 1 1 0 0 0 1 0 0 0 1 0 1 1 1 0 1 0 0 0 0 11 1 1 0 0 1 0 1 1 0 0 0 0 0 1 0 1 0 0 0 1 1 0 11 1 1 1 1 0 0 1 1 1 1 0 1 1 0 1 0 1 0 1 1 1 0 10 1 1 0 1 1 0 0 1 1 0 1 0 0 1 1 0 1 1 0 1 1 1 01 1 0 0 0 1 0 0 1 0 0 1 0 1 1 0 1 1 1 1 0 0 0 11 1 0 0 1 0 0 1 0 1 0 0 0 0 1 0 1 1 1 0 0 0 1 10 0 0 0 0 1 1 0 1 0 0 0 0 1 0 0 0 0 0 0 1 0 0 11 1 1 1 0 0 0 0 0 0 0 1 1 1 0 1 0 1 0 0 0 0 0 00 0 1 0 1 1 0 0 1 1 1 1 1 1 1 1 1 0 0 1 0 0 0 10 1 0 1 0 1 0 1 1 1 0 0 0 0 1 1 0 0 1 1 0 0 1 01 1 1 1 1 0 1 1 1 0 1 1 1 0 0 1 0 1 0 1 1 0 1 00 1 0 1 1 1 0 0 1 1 0 1 1 0 0 0 1 1 0 1 0 0 0 1

Figure 12-5. Bit scrambler sequence

12.8.2 In the case of the current example, the payload consists of the header plus the RS-encoded data as shown in Figure 12-6. When the payload and the scrambling sequences are added together, the result is shown in

- 56 -Figure 12-7. Note that, while the original bit sequence (Figure 12-6) has a preponderance of ZEROs (117 ONEs and 483 ZEROs), the scrambled sequence (Figure 12-7) seems to be more “random” (302 ONEs and 298 ZEROs).

0 1 1 0 0 0 0 0 0 1 0 0 1 0 1 0 1 0 0 0 0 1 0 11 1 1 0 1 1 0 1 0 0 1 1 0 0 1 0 1 1 1 1 0 0 0 11 0 0 0 0 0 0 1 0 0 0 0 0 0 0 0 0 1 0 1 1 0 0 00 1 0 0 0 0 0 1 0 0 0 0 1 1 1 1 0 0 0 0 0 0 0 00 1 1 0 1 0 0 0 0 0 0 0 0 0 0 0 1 0 0 0 0 0 0 01 1 1 0 1 0 1 0 0 1 0 0 0 0 0 0 1 1 0 0 0 0 0 01 0 0 0 0 0 0 0 1 1 0 0 0 0 0 0 0 1 0 0 1 0 1 01 0 0 0 0 0 0 0 1 0 1 0 0 0 0 0 1 1 0 0 1 0 1 00 1 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 1 0 1 1 0 10 0 1 0 1 0 1 0 0 1 0 0 0 0 0 0 0 0 0 0 0 0 0 00 0 1 0 1 0 0 0 1 1 0 0 0 0 0 0 0 0 0 0 0 0 0 00 0 1 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 00 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 00 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 00 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 00 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 00 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 00 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 00 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 00 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 00 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 00 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 1 1 0 1 0 1 00 1 0 0 0 1 1 0 1 0 1 1 0 0 1 0 1 0 0 1 1 1 1 10 0 0 1 1 1 0 0 1 1 1 0 1 1 1 0 1 0 1 1 0 0 1 10 0 1 0 0 1 0 1 1 1 1 0 1 0 1 1 1 1 1 1 0 1 0 0

Figure 12-6. Payload Sequence

0 1 1 1 0 0 1 1 0 1 0 1 0 0 0 1 0 1 0 0 0 0 0 11 1 0 0 1 0 0 0 0 0 1 1 1 1 0 1 0 1 1 1 1 1 0 11 0 0 1 0 1 0 0 1 1 1 0 1 1 1 1 1 0 0 1 0 1 0 10 0 1 0 1 0 1 1 1 1 1 0 0 0 1 1 1 0 0 1 1 0 0 10 0 0 0 0 1 1 0 0 0 1 0 0 0 1 1 1 1 1 0 1 0 0 01 0 0 1 0 0 0 1 0 0 1 0 0 0 0 0 0 1 1 0 0 1 0 00 0 0 0 0 0 0 1 0 1 0 0 1 1 1 0 0 1 0 0 1 0 0 01 0 0 1 0 1 1 1 0 1 0 1 1 0 0 0 1 1 1 1 1 1 1 10 0 0 1 1 1 1 1 1 0 1 1 0 0 1 1 0 0 0 0 0 1 1 10 1 1 0 1 1 1 0 0 0 1 0 0 1 1 1 0 0 0 0 1 1 1 10 1 0 1 1 1 0 0 1 1 0 1 0 1 0 0 1 0 1 1 0 0 0 00 0 0 1 0 0 0 1 1 0 1 1 1 1 1 1 1 0 1 1 1 1 0 11 0 1 0 1 0 1 0 0 1 0 1 0 0 1 0 0 1 1 0 0 1 1 10 0 1 1 1 0 0 0 1 0 0 0 1 0 1 1 1 0 1 0 0 0 0 1

- 57 -

1 1 1 0 0 1 0 1 1 0 0 0 0 0 1 0 1 0 0 0 1 1 0 11 1 1 1 1 0 0 1 1 1 1 0 1 1 0 1 0 1 0 1 1 1 0 10 1 1 0 1 1 0 0 1 1 0 1 0 0 1 1 0 1 1 0 1 1 1 01 1 0 0 0 1 0 0 1 0 0 1 0 1 1 0 1 1 1 1 0 0 0 11 1 0 0 1 0 0 1 0 1 0 0 0 0 1 0 1 1 1 0 0 0 1 10 0 0 0 0 1 1 0 1 0 0 0 0 1 0 0 0 0 0 0 1 0 0 11 1 1 1 0 0 0 0 0 0 0 1 1 1 0 1 0 1 0 0 0 0 0 00 0 1 0 1 1 0 0 1 1 1 1 1 1 1 1 1 1 1 1 1 0 1 10 0 0 1 0 0 1 1 0 1 1 1 0 0 0 1 1 0 1 0 1 1 0 11 1 1 0 0 1 1 1 0 1 0 1 0 1 1 1 1 1 1 0 1 0 0 10 1 1 1 1 0 0 1 0 0 1 1 0 0 1 1 0 0 1 0 0 1 0 1

Figure 12-7. Scrambled Payload

Figure 12-6. Payload sequence

Figure 12-7. Scrambled payload

12.7 9 PHASE CHANGE SEQUENCE

The final sequence of the previous section is nearly ready to be translated into a sequence of phase changes for the D8PSK modulator. However, prior to that step the power stabilization and synchronization sequences must be added to the head of the transmitted bit stream. When this is done, the result is as shown in Figure 12-8.

0 0 0 0 0 0 0 0 00 0 0 1 1 1 0 1 1 0 1 0 0 0 0 1 0 0 0 0 1 0 1 01 0 0 1 0 1 0 1 1 1 1 0 0 0 1 1 1 0 1 0 1 1 1 10 1 1 1 0 0 1 1 0 1 0 1 0 0 0 1 0 1 0 0 0 0 0 11 1 0 0 1 0 0 0 0 0 1 1 1 1 0 1 0 1 1 1 1 1 0 11 0 0 1 0 1 0 0 1 1 1 0 1 1 1 1 1 0 0 1 0 1 0 10 0 1 0 1 0 1 1 1 1 1 0 0 0 1 1 1 0 0 1 1 0 0 10 0 0 0 0 1 1 0 0 0 1 0 0 0 1 1 1 1 1 0 1 0 0 01 0 0 1 0 0 0 1 0 0 1 0 0 0 0 0 0 1 1 0 0 1 0 00 0 0 0 0 0 0 1 0 1 0 0 1 1 1 0 0 1 0 0 1 0 0 01 0 0 1 0 1 1 1 0 1 0 1 1 0 0 0 1 1 1 1 1 1 1 10 0 0 1 1 1 1 1 1 0 1 1 0 0 1 1 0 0 0 0 0 1 1 10 1 1 0 1 1 1 0 0 0 1 0 0 1 1 1 0 0 0 0 1 1 1 10 1 0 1 1 1 0 0 1 1 0 1 0 1 0 0 1 0 1 1 0 0 0 00 0 0 1 0 0 0 1 1 0 1 1 1 1 1 1 1 0 1 1 1 1 0 11 0 1 0 1 0 1 0 0 1 0 1 0 0 1 0 0 1 1 0 0 1 1 10 0 1 1 1 0 0 0 1 0 0 0 1 0 1 1 1 0 1 0 0 0 0 11 1 1 0 0 1 0 1 1 0 0 0 0 0 1 0 1 0 0 0 1 1 0 11 1 1 1 1 0 0 1 1 1 1 0 1 1 0 1 0 1 0 1 1 1 0 10 1 1 0 1 1 0 0 1 1 0 1 0 0 1 1 0 1 1 0 1 1 1 01 1 0 0 0 1 0 0 1 0 0 1 0 1 1 0 1 1 1 1 0 0 0 11 1 0 0 1 0 0 1 0 1 0 0 0 0 1 0 1 1 1 0 0 0 1 10 0 0 0 0 1 1 0 1 0 0 0 0 1 0 0 0 0 0 0 1 0 0 11 1 1 1 0 0 0 0 0 0 0 1 1 1 0 1 0 1 0 0 0 0 0 00 0 1 0 1 1 0 0 1 1 1 1 1 1 1 1 1 1 1 1 1 0 1 10 0 0 1 0 0 1 1 0 1 1 1 0 0 0 1 1 0 1 0 1 1 0 11 1 1 0 0 1 1 1 0 1 0 1 0 1 1 1 1 1 1 0 1 0 0 10 1 1 1 1 0 0 1 0 0 1 1 0 0 1 1 0 0 1 0 0 1 0 1

Figure 12-8. Bit sequence before phase change

This sequence is finally in a form which can be used to generate a sequence of phase changes. This is done by dividing the bit sequence in groups of three (e.g. yi(1), yi(2), yi(3)) and using a look-up table which maps

3-bit symbols into phase changes. For convenience the phase changes are described in terms of a sequence of xi as follows:

where the angle change, , is measured in radians. In this notation, the table is given by Table 12-1. Note that this table can also be expressed as a simple equation. Using the same notation, xi is given by

where the symbol represents modulo 2 summation and the symbol + represents ordinary summation.

Table 12-1. Data encoding

yi(1) yi(2) yi(3) xi

0 0 0 00 0 1 10 1 1 20 1 0 31 1 0 41 1 1 51 0 1 61 0 0 7

Figure 12-9 shows the sequence of phase changes in this notation. The first row is the power stabilization component. The next two rows comprise the synchronization sequence. The next row is the header and the remaining 24 rows are the RS data word.

0 0 00 5 2 3 0 7 1 37 6 2 4 1 4 6 52 7 4 6 0 6 0 14 3 0 2 4 6 5 67 6 1 4 5 4 3 61 3 5 4 1 4 2 10 1 7 3 1 5 6 07 7 3 3 0 1 7 70 0 3 7 5 1 1 07 6 4 6 7 2 5 50 5 5 2 1 7 0 52 2 7 3 2 7 1 53 5 1 6 3 3 4 00 7 2 2 5 4 5 66 3 7 6 1 1 7 51 4 1 0 6 4 7 15 1 2 0 1 3 1 65 4 2 4 4 6 2 62 2 1 6 1 6 6 44 1 1 1 2 2 4 14 3 3 7 1 2 7 20 1 6 0 3 0 1 15 7 0 1 4 6 0 01 2 1 5 5 5 5 20 7 4 5 0 4 6 65 1 4 6 2 5 6 12 4 3 2 1 7 7 6

Figure 12-9. Phase change sequence

12.108 PHASE SEQUENCE

The sequence of phase changes derived in the previous paragraph can be converted into a sequence of phases by summing all the xi up to a particular point. The final sequence is given by Figure 12-10 (assuming that the starting phase is zero radians). If the numbers in the figure are labelled z i, then the sequence of phases is given by

where the angle, , is given in radians. This sequence defines the trajectory of the complex waveform as it passes through the space of I and Q. In other words, these angles define specific sequences of (I, Q) pairs, which are passed through appropriate Nyquist filters to generate the waveform that is sent to the radio’s modulator.

0 0 00 5 7 2 2 1 2 54 2 4 0 1 5 3 02 1 5 3 3 1 1 26 1 1 3 7 5 2 07 5 6 2 7 3 6 45 0 5 1 2 6 0 11 2 1 4 5 2 0 07 6 1 4 4 5 4 33 3 6 5 2 3 4 43 1 5 3 2 4 1 66 3 0 2 3 2 2 71 3 2 5 7 6 7 47 4 5 3 6 1 5 55 4 6 0 5 1 6 42 5 4 2 3 4 3 01 5 6 6 4 0 7 05 6 0 0 1 4 5 30 4 6 2 6 4 6 46 0 1 7 0 6 4 04 5 6 7 1 3 7 04 7 2 1 2 4 3 55 6 4 4 7 7 0 16 5 5 6 2 0 0 01 3 4 1 6 3 0 22 1 5 2 2 6 4 27 0 4 2 4 1 7 02 6 1 3 4 3 2 0

Figure 12-10. Final Phase Sequence

Rob Morgenstern, 06/14/05,

Word conversion failed at this point. I cannot determine what the original text was, although I suspect little change to the text. The contents of the Figure12-10 should be updated.

Part II – Detailed Technical Specifications

1. DEFINITIONS ANDSYSTEM CAPABILITIES

The very high frequency (VHF) digital link (VDL) provides both voice and data service capability operating together or independently. The data capability is a constituent mobile subnetwork of the aeronautical telecommunication network (ATN). In addition, the VDL may provide non-ATN functions. The Standards and Recommended Practices (SARPs) for the VDL are defined in Annex 10, Volume III, Part I, Chapter 6. The SARPs contain introductory and general regulatory information. Additionally, the SARPs include detailed standards of the physical layer protocols and services. This manual contains detailed specifications of link layer protocols and services, subnetwork layer protocols and services, VDL mobile subnetwork dependent convergence function (SNDCF), and the VDL Mode 3 voice unit.

Note.— Where appropriate, references to Annex 10, Volume III, Part I, Chapter 6 have been included in these specifications.

1.1 Definitions

Aeronautical telecommunication network. An internetwork architecture that allows ground, air-ground, and aircraft data subnetworks to interoperate by adopting common interface services and protocols based on the International Organization for Standardization (ISO) Open Systems Interconnection (OSI) Reference Model.

Aircraft address. A unique combination of 24 bits available for assignment to an aircraft for the purpose of air -ground communications, navigation and surveillance.

Asynchronous balanced mode. A balanced operational mode in which a data link connection has been established between two service access points. Either data link entity can send commands at any time and initiate responses without receiving permission from the peer data link entity on the connection.

Asynchronous disconnected mode. A balanced non-operational mode in which no logical data link connection exists between two link layer entities. A connection must be established before data can be sent.

ATN router. An intermediate system used to interconnect subnetworks conforming to the lower three layers of the OSI reference model.

Autotune function. The function, performed by the link management entity, allows a ground station to command an aircraft to change frequencies.

Basic voice service. The voice service that does not use discretely addressable Local User IDs for signalling between the ground station and aircraft stations. This service is available to aircraft stations in TS1, TS2, or TS3 timing state.

Broadcast. A transmission intended to be received by all stations.

Broadcast handoff. The process by which a ground LME commands certain aircraft to execute a link handoff and optionally maintain its current subnetwork connections, without the need to explicitly confirm the link handoff or optionally the subnetwork connection maintenance.

Broadcast link handoff. The process by which a ground LME commands certain aircraft to execute a link handoff to a specific ground station without the need to explicitly confirm the link handoff.


- 65 –

Broadcast subnetwork connection handoff. The process by which a ground LME commands certain aircraft to execute a link handoff to a specific ground station and maintain its current subnetwork connections without the need to explicitly confirm the link handoff or the subnetwork connection maintenance.

Burst. A time-defined, contiguous set of one or more related signal units which may convey user information and protocols, signalling, and any necessary preamble.

Coast timing. A timing state where the aircraft radio is not receiving timing synchronization pulses from the ground station. The aircraft user will attempt to synchronize with any non-coasting aircraft users instead.

Code rate. The ratio of information bits to overall transmitted bits for an error correction code.

Current link (or current ground station). Either the ground-to-aircraft link or the active link when in the process of a handoff.

Data circuit-terminating equipment (DCE). A DCE is a network provider equipment used to facilitate communications between DTEs.

Data link entity (DLE). A protocol state machine capable of setting up and managing a single data link connection.

Data link service (DLS) sub-layer. The sub-layer that resides above the MAC sub-layer. The DLS manages the transmit queue, creates and destroys DLEs for connection-oriented communications, provides facilities for the LME to manage the DLS, and provides facilities for connectionless communications.

Data terminal equipment (DTE). A DTE is an endpoint of a subnetwork connection.

Effective data rate. The actual instantaneous data throughput realized after overheads imposed by bit stuffing and by any forward error correction encoding, but not retransmissions.

Enhanced voice service. The voice service that provides additional operational features over Basic Voice service relying on signalling between the ground station and aircraft stations using Local User IDs for aircraft station addressing. These features include Next Channel Uplink, Aircraft Logged In, Aircraft Transmit ID, Call Queuing Request, and Urgent Downlink Request. This service is only available to aircraft stations who have successfully completed the Net Entry procedures and are in TS1 timing state. The ground station has the option not to support certain Enhanced Voice features through the Supported Options message signalling during the net entry process.

Expedited subnetwork connection establishment. The process by which an aircraft DTE establishes a subnetwork connection with a ground DTE with which it does not have a subnetwork connection during link establishment (or aircraft-initiated handoff) by inserting the CALL REQUEST packet and its response in the link establishment (or aircraft-initiated handoff) frame and its response.

Expedited subnetwork connection maintenance. The process by which an aircraft or ground DTE maintains a subnetwork connection with a DTE with which it has a subnetwork connection during link handoff by inserting the CALL REQUEST packet and its response in the link handoff frame and its response.

Explicit subnetwork connection establishment. The process by which an aircraft DTE establishes a subnetwork connection with a ground DTE with which it does not have a subnetwork connection only after completing the link establishment (or handoff).

Explicit subnetwork connection maintenance. The process by which an aircraft DTE maintains a subnetwork connection with a ground DTE with which it has a subnetwork connection only after completing the link handoff.

Extended Golay code. An error correction code capable of correcting multiple bit errors.

Frame. The link layer frame is composed of a sequence of address, control, FCS and information fields.

- 66 –

Ground network interface. The ground equipment that interfaces with the voice switch, ATN router and the remote radio via the VDL protocols.

Initiated handoff. The transmission process by which a station initiates link handoff.

Internetworking protocol. A protocol that transfers data packets between intermediate systems (IS) and end systems (ES) interconnected by subnetworks and that is supported by the routing protocols and addressing plan.

Link. A link connects an aircraft DLE and a ground DLE and is uniquely specified by the combination of aircraft DLS address and the ground DLS address. A different subnetwork entity resides above every link endpoint.

Link establishment. The process by which an aircraft and a ground LME discover each other, determine to communicate with each other, decide upon the communication parameters, create a link and initialize its state before beginning communications.

Link handoff. The process by which peer LMEs, already in communication with each other, create a link between an aircraft and a new ground station before disconnecting the old link between the aircraft and the current ground station.

Link layer. The layer that lies immediately above the physical layer in the Open Systems Interconnection protocol model. The link layer provides for the reliable transfer of information across the physical media. It is subdivided into the data link sub-layer and the media access control sub-layer.

Link management entity (LME). A protocol state machine capable of acquiring, establishing and maintaining a connection to a single peer system. An LME establishes data link and subnetwork connections, “hands-off” those connections, and manages the media access control sub-layer and physical layer. An aircraft LME tracks how well it can communicate with the ground stations of a single ground system. An aircraft VME instantiates an LME for each ground station that it monitors. Similarly, the ground VME instantiates an LME for each aircraft that it monitors. An LME is deleted when communication with the peer system is no longer viable.

Local User ID. A unique identification of a specific aircraft that is used by the participants in a VDL subnetwork. The Local User ID consists of a 2-bit User Group ID prefix followed by a 6-bit numerical suffix.

MAC cycle. The standard timing cycle that consists of two consecutive (even and odd) time division multiple access (TDMA) frames.

M burst. A management channel data block of bits used in VDL Mode 3. This burst contains signalling information needed for media access and link status monitoring.

Media access control (MAC). The sub-layer that acquires the data path and controls the movement of bits over the data path.

Mode 3. A voice and data VDL mode that uses D8PSK modulation and a TDMA media access control scheme.

Multicast. A transmission intended to be received by multiple stations.

Network layer (NL). The layer that provides the upper layers with independence from the data transmission and routing functions used to connect systems. The network layer is responsible for routing and relaying functions both within any subnetwork and throughout the aeronautical internetworking domain.

New link (or new ground station). After successful completion of handoff (or link establishment), the new “current” link.

- 67 –

N(r). The receive sequence number at the link layer, which indicates the sequence number of the next expected frame (and explicitly acknowledges all lesser numbered frames).

N(s). The send sequence number at the link layer, which indicates the sequence number associated with a transmitted frame.

Old link (or old ground station). Following link establishment during a handoff, the link that was previously the “current” link becomes the “old” link.

Physical layer. The lowest level layer in the Open Systems Interconnection protocol model. The physical layer is concerned with the transmission of binary information over the physical medium (e.g. VHF radio).

Polling. A procedure by which the ground station interrogates aircraft stations, one at a time, to determine status/control functions.

Private parameters. The parameters that are contained in exchange identity (XID) frames and that are unique to the VHF digital link environment.

Proposed link (or proposed ground station). The link being negotiated (in a handoff) to replace the current link.

Quality of service (QoS). The information relating to data transfer characteristics used by various communication protocols to achieve various levels of performance for network users.

Reed-Solomon code. An error correction code capable of correcting symbol errors. Since symbol errors are collections of bits, these codes provide good burst error correction capabilities.

Requested handoff. The one-transmission process by which a station requests its peer entity to initiate a link handoff.

Service primitives. The status and control information that must be available to the receiving entity to properly process incoming information. A service primitive may contain parameters. If parameters exist, they describe information that is defined either as mandatory (M) or optional (O) for conformance to a particular communications standard.

Service provider. An entity at a layer that provides services to the layer above. These services are provided at service access points through the use of service primitives.

Service user. An entity at a layer that makes use of the services that are provided at service access points by the layer below through the use of service primitives.

Slotted aloha. A random access strategy whereby multiple users access the same communications channel independently, but each communication must be confined to a fixed time slot. The same timing slot structure is known to all users, but there is no other coordination between the users..

Squelch window. The time period a radio searches for the beginning of a message.

Subnetwork connection. A long-term association between an aircraft DTE and a ground DTE using successive virtual calls to maintain context across link handoff.

Subnetwork connection maintenance. The process by which the VDL SNDCF maintains subnetwork context from one subnetwork connection to the next during handoffs.

Subnetwork connection management. The process by which the VDL SNDCF initially establishes a connection and then maintains it during handoffs.

- 68 –

Subnetwork dependent convergence function (SNDCF). A function that matches the characteristics and services of a particular subnetwork to those characteristics and services required by the internetwork facility.

Subnetwork entity. In this document, the phrase “ground DCE” will be used for the subnetwork entity in a ground station communicating with an aircraft; the phrase “ground DTE” will be used for the subnetwork entity in a ground router communicating with an aircraft station; and, the phrase “aircraft DTE” will be used for the subnetwork entity in an aircraft communicating with the station. A subnetwork entity is a packet layer entity as defined in ISO 8208.

Subnetwork layer. The layer that establishes, manages and terminates connections across a subnetwork.

Switching time. In VDL Mode 3, the switching time is the time for a radio to switch from reception to transmission (R/T), or from transmission to reception (T/R). This time is measured from the end of the ramp-down of one burst (i.e. when its power is down from the nominal level of – 20 dBc) until the beginning of the ramp-up of the next burst (i.e. when the power is down from the nominal level of – 40 dBc.)

System. A VDL-capable entity. A system comprises one or more stations and the associated VDL management entity. A system may either be an aircraft system or a ground system.

System configuration. Defines the allocation of TDMA time slot resources to various user groups supported by ground stations sharing the same 25 kHz VHF channel.

T. The symbol period or 1/symbol rate.

TDMA frame. The basic unit of time (120 ms) which consists of 3 or 4 time slots. In any voice communication, a radio will transmit digitized voice bits periodically once per frame.

Time slot. A TDMA timing unit allocated for an M burst and V/D burst.

Truncation. A timing state where the aircraft timing uncertainty is such that the V/D (voice) burst duration is reduced to increase the guard time. The additional guard time provides more room for timing error without interfering with adjacent time slots.

Unicast. A transmission addressed to a single station.

User group. A group of ground and/or aircraft stations which share voice and/or data connectivity. For voice communications all members of a user group can access all communications. For data, communications include point-to-point connectivity for air-to-ground messages, and point-to-point and broadcast connectivity for ground-to-air messages.

User Group ID. User Group ID, which is synonymous with Group ID, is a 2-bit code that together with the channel frequency uniquely identifies the VDL Mode 3 communications channel.

V/D burst. A burst that is used for the transmission of a user’s voice or data. V/D (voice) is a V/D burst used for voice transmission and V/D (data) is associated with data transmission.

VDL management entity (VME). A VDL-specific entity that provides the quality of service requested by the ATN-defined SN-SME. A VME uses the LMEs (that it creates and destroys) to enquire the quality of service available from peer systems.

VDL station. An aircraft-based or ground-based physical entity, capable of VDL Mode 2, 3 or 4.

Vocoder. A low bit rate voice encoder/decoder.

- 69 –

Vocoder frame. A window of time in which the analog speech waveform is used by the vocoder to process and generate the encoded digital voice output and the reverse process to regenerate, from the digital voice data block, the corresponding analog speech waveform.

Voice analysis. The process where speech is modelled and converted into parameters.

Voice synthesis. The process where voice analysis parameters are used to generate speech.

Voice unit. A device that provides a transparent, simplex audio and signalling interface between the user and VDL.

1.2 Radio channels andfunctional channels

Note. — The radio frequency range for the aircraft station and ground station are standardized in the VDL SARPs, Annex 10, Volume III, Part I, Chapter 6.

1.3 System capabilities

Note. — The VDL communication functions must meet the general requirements standardized in the VDL SARPs, Annex 10, Volume III, Part I, Chapter 6. Guidance material for the VDL is included as an attachment to these Technical Specifications.

1.4 Air-ground VHF digitallink communicationssystem characteristics

Note.— The characteristics of the air-ground VDL communications system used in the international aeronautical mobile service must be in conformity with the standards of the VDL SARPs, Annex 10, Volume III, Part I, Chapter 6. This specification includes definition of radio frequency bands, channel spacing and emissions polarization.

2. SYSTEM CHARACTERISTICS OFTHE GROUND INSTALLATION

Note.— Annex 10, Volume III, Part I, Chapter 6 contains the SARPs relating to: – the radio frequency stability of VDL ground station equipment;– the recommended effective radiated power of the VDL ground station equipment;– spurious emissions standards for the VDL ground station equipment; and– adjacent channel emissions standards for the VDL ground installation.

3. SYSTEM CHARACTERISTICS OF THE AIRCRAFT INSTALLATION

Note.— Annex 10, Volume III, Part I, Chapter 6 contains the SARPs relating to: – the radio frequency stability of VDL aircraft station equipment;– the effective radiated power of the VDL aircraft station equipment;– spurious emissions standards for the VDL aircraft station equipment; – adjacent channel emissions standards for the VDL aircraft installation; and– specified error rates, receiving function sensitivity, undesired signal rejection, and interference immunity

standards for the VDL aircraft station.

- 70 –

4. PHYSICAL LAYERPROTOCOLS AND SERVICES

4.1 Functions

Note.— The general functions provided by the VDL physical layer are standardized in the VDL SARPs, Annex 10, Volume III, Part I, Chapter 6.

4.2 Mode 3 physical layer

Note. — The standards of the VDL Mode 2 and VDL Mode 3 physical layer are provided in the VDL SARPs, Annex 10, Volume III, Part I, Chapter 6.

5. LINK LAYER PROTOCOLSAND SERVICES

5.1 General information

Note.— General information regarding the link layer protocols including functionality and services is standardized in the VDL SARPs, Annex 10, Volume III, Part I, Chapter 6. The VDL Mode 3 media access control (MAC) sub -layer, data link service sub-layer, and link management entity (LME) sub-layer are specified in 5.2, 5.3 and 5.4, respectively, of these Technical Specifications.

5.2 Mode 3 MAC sub-layer

The MAC sub-layer shall support both voice and data operation. For data operation, the MAC sub-layer shall provide for the transparent acquisition of the shared communication path. For voice operation, the MAC sub-layer shall support simplex voice communications with pre-emptive access for authorized ground users.

5.2.1 MAC services for Mode 3

5.2.1.1Multiple access. For data operation, the MAC sub-layer shall implement a ground station-centralized, reservation-based access to the channel which permits priority-based access. The sub-layer shall implement both a polling-based access and a random access method for reservation requests.

Note 1.— These methods provide a fair access (i.e. non-bias) for all aircraft.

Note 2.—

Uplink data transmission is scheduled directly at the ground station without a reservation request.

For voice operation, access shall be primarily supported by the user based on a “listen before push-to-talk” discipline. Additionally, authorized ground users shall have the capability to pre-empt aircraft users occupying the voice channel. Aircraft stations shall not have pre-emption capability.

Note 23.— This pre-emptive access provides a limited capability to cope with a “stuck” microphone situation by allowing authorized ground users to deactivate the transmitter of the offending user.

- 71 –

5.2.1.2Channel occupancy.

Note 1.— Channel occupancy is the amount of time spent a transmitter is allowed to transmit on the channel per access event.

5.2.2.1.2Time slot. A TDMA time slot shall be of 30 or 40 ms duration depending on the System Configuration in effect for the ground station. A time slot shall be the basic channel resource allocated to a user group for either voice or data transmission.

Note 1.— Multiple time slots can be assigned to a single user group (e.g. voice and data).

Note 2.—Use of 30 ms slots is considered Normal Range operation; use of 40 ms slots is considered Long Range operation.

5.2.2.1.3Bursts. A TDMA time slot shall consist of either a) one Management (M) burst and one Voice or Data (V/D) burst, b) four M bursts, or c) one M burst and one handoff check message (H) burst. The format of the bursts shall be as specified in Figures 5-1a)1, 5-1b) and 5-1c). Figure 5-1a) shows a typical time slot to consist of an M burst followed by a V/D burst. Figure 5-1b) shows slot A of a 3T configuration downlink to consist of four management M bursts. Figure 5-1c) shows slot A of a 3T configuration uplink to consist of a management burst and an H burst.

Note 1.— An M burst is used for signalling and link management. A V/D burst is used for user information. An H burst is used to facilitate automated handoff of ground stations.

The timescale hierarchy for TDMA frames, time slots and bursts shall be as shown in Figures 5-2a), 5-2b) and 5-2c). Figure 5-2a) shows the TDMA frame structure of a normal range non-3T configuration. Figure 5-2b) shows the TDMA frame structure of a normal range 3T configuration. Figure 5-2c) shows the TDMA frame structure for long range configurations.

Note 2.— The only difference in the timing structure between the normal range 30 ms slots and the long range 40 ms slots is an increased propagation guard time allowance. The exact allocation of propagation guard time is reflected in Tables 5-1, *2 5-2, and 5-3.

5.2.2.3User groups. An aircraft radio shall be associated with a user group for purposes of channel access for voice and data transmissions and for aircraft addressing functions used by the ground station. A single VDL frequency assignment (i.e. ground station radio) for Mode 3 shall support from 1 to 4 user groups depending on the system configuration in effect for that ground station.

5.2.2.5.1Logical burst access channels. For configurations 4V, 2V2D, 3V, 3V1D, and 2V1D, VDL Mode 3 shall grant media access individually for each of up to 8 separate Logical Burst Access Channels (LBACs) numbered 1-8. A

1 ? All figures are located at the end of these Technical Specifications.

2 ?All tables are located at the end of these Technical Specifications.

- 72 –

description of each LBAC for these cases is given by Table 5-5. For the 3T configuration, VDL Mode 3 shall grant media access individually for each of up to 18 separate LBACs numbered 1-18. A description for each LBAC for the 3T configuration is given by Table 5-2. For the 3S configuration, VDL Mode 3 shall grant media access individually for each of up to 10 separate LBACs numbered 1-10. A description for each LBAC for the 3S configuration is given by Table 5-23a). For the 2S1X configuration, VDL Mode 3 shall grant media access individually for each of up to 7 separate LBACs numbered 1-7. A description for each LBAC for the 2S1X configuration is given by Table 5-23b). For the 1V3D configuration, VDL Mode 3 shall grant media access individually for each of up to 16 separate LBACs numbered 1-16. A description of each LBAC is given in Table 5-2c3d). For the 1V2D configuration, VDL Mode 3 shall grant media access individually for each of up to 12 separate LBACs numbered 1-12. A description of each LBAC is given in Table 5-3e2d).

Note.— MAC cycles for mosteach system configurations are shown labelled with each LBAC in the Implementation Aspects of this manual.

5.2.2.5.2Burst access timing. The system configuration code provided in the M uplink burst (Appendix B, 2.1) shall be used by the aircraft stationradio in conjunction with the timing reference point (TRP) to establish proper burst timing. For configurations 4V, 2V2D, 3V, 3V1D, and 2V1D, LBACs shall be established according to Table 5-1; for the 3T configuration, LBACs shall be established according to Table 5-2; for the 3S configuration, LBACs shall be established according to Table 5-23a); for the 2S1X configuration, LBACs shall be established according to Table 5-23b); for the 1V3D configuration, LBACs shall be established according to Table 5-2c3d); for the 1V2D configuration, LBACs shall be established according to Table 5-2d3e). The timing reference point (TRP) shall be established per 5.2.4.1. The relative timings among the TRPs of different user groups for the various system configurations are shown in Table 5-3c).

5.2.3.1Parameter t (truncation). This parameter shall be the number of MAC cycles which can elapse in normal timing state 1 (TS1) while not receiving a primary time reference (defined in 5.2.4.1.1) without requiring transmitted V/D (voice) bursts to be truncated.

Note.— This parameter is determined by the symbol clock stability of the aircraft VDL radio and represents the number of MAC cycles prior to accumulating a timing drift of ± 1 symbol period.

5.2.3.1.1Validity window. A timing beacon (with appropriate Slot Beacon ID and Ground Station Code) shall be considered valid (i.e., acceptable for aircraft radio timing updates) if it is received within the validity window with respect to the previous beacon’s timing reference. The validity window shall be defined to extend from –K(n+1)+1 to +1 symbol periods, where K=4 for 4-slot configurations and K=8 for 3-slot configurations and n is the squelch window parameter. To allow and aircraft radio to adapt in a timely manner to an abrupt change in the location of a beacon transmitter, an aircraft radio shall extend its validity window to include all relative times whenever its CTC1 counter exceeds 24.

Note.— This definition takes into account the 1 symbol period accuracy assumed for the aircraft radio timing. It is also designed to support diversity site and selective keying operation, where it is desirable that the aircraft radio should base its timing on the signal from the closest ground site.

5.2.3.1.2Receipt of Beacon Information. Aircraft radios shall decode and process information contained in any ground beacon signal with BeaconSlot ID and Ground Station Code appropriate for the aircraft radio’s current net.

Note.— Although an aircraft radio must use the validity window to restrict which ground beacons it should use to update its timing, it should accept and react to any ground beacon associated with its net. This allows an aircraft radio to have its timing based on the closest ground site, while getting system information in the most timely manner.

- 73 –

5.2.3.2Parameter f (free running). This parameter shall represent the maximum Coast Timing Counter 3 (CTC3) (5.2.4.1.3) value beyond which the free running timing state 3 (TS3) is entered. This parameter shall represent the maximum number of MAC cycles of “coasting” which must elapse without receipt of a primary timing or alternate timing signal (ALTS) before transitioning to TS3. ALTS is defined in 5.2.4.1.2.

Note.— This parameter represents at least two whole polling cycles.

5.2.4 Description of procedures

Media access shall will observe the procedures described in this Section in addition to observing the timing structure limitations of 5.2.2.1.

5.2.4.1 Timing acquisition and maintenance

5.2.4.1.1Primary timing reference. The aircraft stationradio shall derive its primary timing from the received M uplink burst (LBAC 11 for 3T configuration, either LBAC 5, 7 or 9 for 3S configuration, either LBAC 4 or 6 for 2S1X, LBAC 9 for 1V3D, LBAC 7 for 1V2D, and LBAC 5 for other configurations) associated with its assigned user group and Ground Station Code. The timing reference point shall be established based on the centre of the first received symbol of the synchronization sequence at the antenna port of the M uplink burst associated with the user group assigned. VDL Mode 3 media access opportunities for transmitted bursts shall be set relative to this reference in terms of transmitted symbol periods as shown in Tables 5-1, 5-2, and 5-2a, 5-2b, 5-2c and 5-2d3.

Note 1.— To remain in the normal timing state (TS1), an aircraft stationradio should maintain its primary timing to within ± 1 symbol period of the correct time based on the position of the aircraft relative to the ground station. This timing tolerance is the basis for developing the timing state transition rules of 5.2.4.1.4 and 5.2.4.1.5. Refer to 5.2.3.1 for this timing accuracy requirement.

Note 2.— The relative timings among the TRPs of the different user groups are as shown in Table 5-3c).

5.2.4.1.2Alternate timing reference. When required by the procedures of 5.2.4.1.3 through 5.2.4.1.5, the aircraft stationradio shall derive alternate timing in truncate timing state 2 (TS2) from alternate timing sources (ALT). ALT are poll responses (except in the 3T configuration) associated with the user group or another user group (if available) and M uplink bursts in slots other than the one controlling the aircraft radio’s user group received during that MAC cycle.

5.2.4.1.2.3This alternate timing shall be derived based on the first received symbol of the synchronization sequence of the received poll responses or uplink M bursts. The aircraft radiostation shall accept timing from ALTs if they obey the following rule:

Tr < Ta + 0.04 CTC2 Ts

where Ta is the expected time of an ALT as determined by the aircraft radio’s clock, T r is the actual time of arrival of the ALT with respect to the aircraft radio’s clock, Ts is the symbol period, and CTC2 is the current value of the CTC2 counter (defined in 5.2.4.1.3). If CTC2 is greater than 800, all ALT time updates shall be accepted. If the time is updated, CTC2 shall be reset to 0.

5.2.4.1.3Coast timing counters. The aircraft VDL radio shall maintain Coast Timing Counter 1 (CTC1), Coast Timing Counter 2 (CTC2), and Coast Timing Counter 3 (CTC3) in order to determine the timing state for each MAC cycle. CTC1, CTC2 and CTC3 shall be maintained per Table 5-8a). The maximum value to which each of the three CTC counters is allowed to increment shall be 1023. Once the CTC counter reaches 1023, that CTC counter shall be held at that value indefinitely until reset is required in accordance with the rules states in Table 5-8a).

- 74 –

5.2.4.1.4Timing states. The aircraft VDL radio shall take on one of four possible timing states for non-3T configurations as determined by the CTCs as shown in Table 5-8b). The timing states for 3T configuration are shall be as defined in Table 5-8c).

5.2.4.1.4.1The aircraft VDL radio shall support transmissions of downlink M bursts and V/D (data) bursts only in Normal TS1 timing state. The aircraft VDL radio shall receive uplink M bursts and downlink M bursts in all timing states in order to extract timing information from all possible PTSs and ALTSs.

5.2.4.1.4.2Except in 3T configuration, the aircraft VDL radio shall enter truncate timing state 2 (TS2) from TS1 when the number of MAC cycles, which has elapsed without receiving a primary timing reference, is equal to or exceeds the value of parameter t. Voice operation support during the different timing states is specified in 5.2.4.2.

Note.— The timing state controls operation of the Voice Unit (Section 8 of these Technical Specifications).

5.2.4.1.4.3Recommendation.— An indication of the Transceiver timing state should be provided to the VDL Mode 3 aircraft user.

Note.— The timing state controls operation of the Voice Unit (Section 8 of these Technical Specifications).

5.2.4.1.5Timing state transitions. When in TS1, an aircraft stationradio in non-3T configurations shall transition through TS2 before entering TS3. An aircraft stationradio in 3T configuration shall transition from TS1 through a non-transmit transition period before entering TS3. When in TS3, the aircraft stationradio shall accept time from any ALT and enter TS2. Any aircraft stationradio shall immediately revert to TS1 upon reception of a valid primary timing signal.

5.2.4.1.5.1 During a PTT event, no timing state transition (except for transition to TS0 in response to a new tuning command) shall occur until that PTT event is completed. For data transmissions, a timing state transition shall be delayed (except for transition to TS0 in response to a new tuning command) until the current message transmission is completed. In the event that an aircraft station is engaged in a PTT event and a data event (transmit or receive) simultaneously, the timing state transition shall be delayed until both events are completedTiming state transitions shall be immediate except for the following:.

5.2.4.1.5.2Timing State Transitions other than to TS0: Timing state transition shall be delayed until completion of all V/D burst transmissions for any ongoing PTT event and/or any data event for which the aircraft radio has already received a reservation but has not completed the scheduled V/D (data) transmission.

5.2.4.1.5.3Timing state transition to TS0: Upon receiving a tuning command, the aircraft radio shall initiate the Leaving Net message transmission (if required) and then tune to the next channel and transition to TS0 when the Leaving Net transmission is completed.

Note 1.— In the case of data transmission, the timing state transition other than to TS0 occurs after the completion of the last V/D (data) burst transmission. Specifically, the aircraft radio will transition without waiting for ground ACK.

Note 2.— Section 5.2.4.3.4 does not prevent the aircraft radio from continuing to transmit voice and data until the Leaving Net message transmission is transmitted. But, all voice and data transmissions must be terminated as soon as the Leaving Net message is transmitted.

5.2.4.2Voice operation support. The aircraft VDL stationradio shall grant media access on a MAC cycle-by-MAC cycle basis based on the timing state and the voice signalling information received in the most recent M uplink burst of the previous MAC cyclethat was successfully received prior to transmission of the V/D (voice) burst. (If an M uplink

- 75 –

burst is missed, the information is assumed not to have changed.) The rules for media access shall be as described in Table 5-9. Furthermore, if Table 5-9 indicates 'Access', the MAC sublayer shall determine if there was a V/D (voice) burst in the previous TDMA frame from another user. If so, access shall be granted if the V/D (voice) header indicated an EOM; otherwise, the access shall be denied.

5.2.4.2.1The MAC sublayer shall set the CTS flag for the Voice Unit (Section 8.4.3) when access is allowed; otherwise, the CTS flag is cleared. The setting of the CTS flag for each TDMA frame shall coincide with the MAC timing associated with when the V/D (voice) burst is to be transmitted per Tables 5-1, 5-2, and 5-3 as appropriate. The conditions of both the timing state and the signalling shall be met in order to be granted access in a given MAC cycle. The aircraft station shall process received V/D (voice) bursts with correct Group ID while in timing states TS1 and TS2.

Note 1.— Reception of voice in TS0 is not required. Transmission of voice in TS0 is not permitted.

Note 2.— Reception and transmission of voice in TS3 is specified separately in Section 5.2.4.2.2. TS3 Voice Operations.

5.2.4.2.2In the absence of a valid Voice Signal indication, an aircraft radio shall declare the channel idle in the current slot if the V/D (voice) burst header in the previous TDMA frame is detected with an EOM=1. The channel is declared to be busy if the EOM=0. If the V/D (voice) bursts are also absent, the aircraft radio shall declare channel idle in the current TDMA frame, only if the two V/D (voice) slots were determined to be idle in the two previous TDMA frames.

Note.— In the absence of Voice Signal indications to the contrary, aircraft radios will coast through the loss of a single V/D (voice) burst.

5.2.4.2.31The Voice Unit (Section 8 of these Technical Specifications) shall be notified of the timing state during each MAC cycle.

5.2.4.2.42The maximum duration of an aircraft PTT event shall be limited to no more than 35 seconds (approximately 145 MAC cycles). Aircraft stationradio shall automatically terminate the PTT event upon reaching this limit.

5.2.4.2.53To resolve conflicts that may arise due to nearly simultaneous assertion of the PTT of multiple radios, the following rule shall applyAny aircraft VDL transmitting a voice signal shall cease transmitting if it receives a valid uplink M channel signal with the Voice Signal code set to any value other than 01 2 (occupied). (See 2.1.4 of Appendix B.) This rule shall apply for the duration of the PTT event, starting two cycles after the first voice burst transmission.:

5.2.4.2.5.1If, within the first 2 MAC cycles after beginning to transmit V/D (voice) bursts, the aircraft radio successfully receives one or two uplink Beacons and none indicates “Occupied,” then it shall immediately cease transmitting.

5.2.4.2.5.2 After the first 2 MAC cycles (i.e. Cycles 3 and beyond), the radio transmitting voice shall disregard the voice signaling field unless it indicates 'Ground Access'.

Note 1.— Currently, only the 3T system configuration supports voice reservation signalling.

Note 2.— It is desirable for implementation simplicity to be able to remain in the same timing state for the duration of a PTT event. Putting an upper limit on the duration of an aircraft PTT event makes this practical. This limit does not apply to the ground station because of normal AT operating procedures.

Note 3.— The 35 seconds limit on an aircraft PTT event is based on the Automatic Termination of transmission requirement in RTCA/DO-207, “Minimum operational performance standards for devices that prevent blocked channels used in two-way radio communication due to unintentional transmissions”, January 25, 1991.

- 76 –

5.2.4.2.6User Notification. The aircraft station shall notify the users if any of the following events occurs:

a) Voice transmission (including unintentional transmission) is stopped when received uplink M burst voice signal code is other than 01b,

b) PTT event excees the 35-second limit,c) PTT is asserted while access is not allowed (see Table 5-9).

Note.— The methods used for user notification are beyond the scope of this document. Notification to the user could, as an example, be implemented as signalling to the transceiver external interface or implemented as a short duration audible signal that is mixed with the received audio.

5.2.4.2.57Free Running Voice Operation. While in TS3, the aircraft station radio shall process all V/D (voice) bursts without regard to Group ID. Multiple V/D (voice) bursts with different Group ID shall be processed on a first-come-first-serve basis, regardless of Group ID. Once a particular V/D (voice) burst is processed, the aircraft station radio shall continue to process only V/D (voice) bursts from the same user until EOM is indicated.

5.2.4.2.57.1The aircraft station radio in TS3 timing state shall maintain system timing autonomously. While in TS3, the aircraft station radio shall transmit V/D (voice) bursts in the assigned time slot based on the system timing maintained by the aircraft.

Note.— Without beacons and ALT available to correct its time, an aircraft radio in TS3 is expected to drift relative to UTC at a rate determined by the stability of its reference oscillator.

5.2.4.3 Link management support

5.2.4.3.1Polling. When directed by the ground LME, the ground MAC sub-layer shall transmit a Poll Request message. An uplink beacon (see Table B-1) shall be used to poll a particular aircraft by including its Local ID. If an uplink beacon is not intended to poll any user, the Aircraft ID shall be set to zero. Whenever a beacon is transmitted with the Aircraft ID field set to zero, the LBAC which would otherwise be reserved for a Poll Response transmission becomes available for random access (RA) transmissions (see Section 5.2.2.5.3). The aircraft MAC sub-layer shall deliver a received Poll Request message that is addressed to it to the aircraft LME. Upon receiving a Poll Response from the aircraft LME, the aircraft MAC sub-layer shall stop the Reservation Request retransmission using the random access M-channel until a new request is generated. The aircraft station radio shall respond to a poll by the ground station by transmitting a Reservation Request message in LBAC 2, 3 or 4 for 3T configuration, and LBAC 1 for other configurations in the MAC cycle immediately following the poll. This opportunity to request a reservation shall be consistent with the procedures of 5.2.4.4.3.3.3. The Number of Slots Requested field and the Voice Request field shall be used independently for data access and voice request, respectively. If the aircraft station does not need to send data reservation or voice request information, the Number of Slots Requested field and the Voice Request field shall be set to zeroIf no reservation is needed by the aircraft VDL, the Number of Slots Requested field and the Voice Request field shall be set to zero.

5.2.4.3.1.1The ground MAC sub-layer shall deliver a received Poll Response message to the ground LME. The ground station shall respond to a Reservation Request message contained in a Poll Response using the procedures identified in 5.2.4.4.3.3.1 through and 5.2.4.4.3.3.3. Except during the net entry process (see Section 5.4.2.1.2), Nno response shall be sent by the ground station if the number of slots in the reservation field is zero and the Voice Request field is encoded as “No Request”.

5.2.4.3.2.1Transmission attempts for the Net Entry Request message shall be based on the algorithm of Appendix D for selecting an appropriate downlink M-burst LBAC to transmit the Net Entry Request message. The parameter RE (defined in 5.2.3.5) is a MAC Service System Parameter. The value of RE is given in Table 5-7b). This procedure shall

- 77 –

be terminated when a valid Net Entry Response message is received or the number of retransmissions exceeds NL1 (see Table 5-10a)).

5.2.4.3.3.4In the case that the aircraft station has both a Reservation Request message and a Next Net ACK message to transmit when polled, the Next Net ACK message shall be transmitted. The Next Net ACK message format is shall be as defined in Appendix B, 2.13.

5.2.4.3.4Leaving net message. When directed by the LME (5.4), the aircraft MAC sub-layer shall select an appropriate downlink M-burst LBAC to transmit a Leaving Net message based on the algorithm of Appendix D. There shall be no retransmission of a Leaving Net message.

5.2.4.3.4.1The aircraft station radio shall stop all transmitting and receiving activities on that net after the aircraft MAC sublayer sends the Leaving Net message.

5.2.4.3.4.2Tuning to Next Net. Upon completion of the Leaving Net message transmission, the aircraft LME shall automatically tune to the next net, provided that the aircraft LME has received the Next Net information.

Note 1.— It is not necessary for the aircraft to acknowledge the Next Net message to initiate Leaving Net message transmission.

Note 2.— See RTCA, DO-224A, Section 3.7.2 for tuning performance requirements.

Section 5.2.4.3.7 Deleted

5.2.4.3.7Mutiple Radio Installations. If multiple radios are installed on an aircraft station and are operating on the same channel, the radio systems shall coordinate their MAC sublayers, as necessary, so that only one radio acknowledges to polls and next net messages, and that one radio acknowledges uplink V/D (data) uplink bursts.

Note.— This requirement is to resolve issues with multiple radios tuned to the same net, or for system configurations 3V1D/2V1D with multiple radios sharing the same data slot regardless of different local identifiers. Arbitration of downlink messaging in these cases is already handled by the router.

5.2.4.4.3.3.4The rules described in Section 5.2.4.4.3.3.1 on sending a new reservation request after the aircraft has received a Reservation Response message with slot assignment for the outstanding request shall apply when the outstanding request is via Poll Response.

Note.— The difference between replacing an outstanding Reservation Request through Poll Response and Slotted Aloha Random Access is that more time slots can be requested in Poll Response, while the number of time slots of the new request has to be equal to or less than the outstanding request in the case of random access.

5.2.4.4.3.4Reservation request retransmission. The aircraft radio shall select a downlink M-burst LBAC for random access Reservation Request Retransmission based on the process of Appendix D and the random number generator algorithm of Appendix C. The Reservation Request Retransmission shall be repeated until one of the following occurs: (1) a valid response is received, (2) a poll is received, or (3) the number of retransmissions exceeds NM1 (defined in Section 5.2.3.3). The value for NM1 is given in Table 5-7a). When the number of reservation request retransmissions exceeds NM1, random access reservation request retransmission shall be terminated. After the random access

- 78 –

reservation request retransmissions have been terminated, the aircraft station radio shall continue to provide the data reservation in poll responses.

5.2.4.4.3.5Reservation confirmation indication. Upon receipt of a Reservation Response (Appendix B, 2.2.3) signalling an “access scheduled,” the MAC sub-layer shall authorize transmission of all segments of the user message starting in the slot indicated.

5.2.4.4.3.6Acknowledgement protocol. Acknowledgements of downlink transmissions sent by the ground station shall be processed by the aircraft DLS to indicate completion of downlink data transfer.

5.2.4.5MAC data frame transmission/retransmission. After each transmission that requires acknowledgement, the MAC sub-layer shall maintain a retransmission timer T1 for each destination. This timer is the time that the MAC sub-layer shall wait for an acknowledgement prior to a new transmission. Timeout of the T1 timer shall result in a request to the DLS for a new frame group. This implies that the VDL shall replace the retransmission of the outstanding frame group with the transmission of a new frame group, if the new frame group has higher priority frames. new frame group shall include all of the frames in the outstanding frame group requiring acknowledgement that are associated with the T1 expiration, with no new acknowledged frames added. The number of successive T1 timer expirations to a specific destination shall not exceed N2 times. The LME shall be informed when the successive T1 timer expirations exceeds N2 times, signalling it to terminate the link to that destination. The N2 counter shall be reset upon the successful acknowledgement of the message.

5.2.4.5.1The maximum transmission delay for an uplink ACK for a VDL Mode 3 ground station shall not exceed T_ack, where T_ack is in MAC cycles. The delay associated with transmitting an uplink ACK is defined from the time the last V/D (data) burst is received to the time that the ACK transmission begins. If the ground station is unable to transmit the ACK within T_ack after the receipt of the last downlink V/D (data) burst, it shall not transmit the ACK. For aircraft stationsradios, the specific downlink M channel for ACK transmission is defined in 5.3.5.2.

5.2.4.5.2The MAC data frame transmission system parameters T1, T_ack, and N2 are defined in Table 5-11.

Note.— Upon T1 timer expiration and depending upon the message priorities of the current message queues, the new frame group generated by the DLS sub-layer to be send to the MAC sub-layer for transmission mustmay not include all the previously transmitted acknowledged frames requiring acknowledgement, which are associated with event of the T1 timer expiration, and possibly new frames that do not require acknowledgement, depending upon the message priorities of the current message queues.

5.2.4.6Automated handoff for 3T configuration. System configuration 3T shall support automated handoff of ground stations for aircraft stationsradios, which are capable of re-tuning frequencies within 2 ms. LBAC 12 shall be broadcast by all participating ground stations permitting an aircraft station radio to listen to all nearby ground stations (one station per MAC cycle) and determine which ground station provides the strongest signal. In the LBAC 12 burst, the current ground station shall broadcast its Ground Station Code, the Ground Subnetwork Address of the GNI attached to the router, and the frequencies of the next four proximate nets to be monitored. The aircraft station radio shall scan through these adjacent frequencies in LBAC 12 and compare signal quality from the ground stations.

5.2.4.6.1 After scanning the last adjacent frequency, the aircraft station radio shall scan the current ground station frequency again, which may contain different adjacent frequency information if there are more than four nets to be monitored. The aircraft station radio shall accumulate and maintain a database of all nearby nets to be scanned, including frequency, Ground Station Code, and their latest signal quality measurement.

5.2.4.6.2 When the signal quality from a proximate ground station is determined to provide a better link than the one provided by the current ground station in accordance with an algorithm selected by the equipment manufacturer, the aircraft station radio shall initiate an automatic handoff to the new ground station. If the aircraft station radio cannot acquire the new net, it shall attempt to acquire the other nets and the previous net, in descending order of signal quality. If the aircraft station cannot locate a new ground station, the LME shall be notified. If at any time no LBAC_ 12 signals are received, the LME shall be notified.

- 79 –

Note 1.— The signal quality comparison should be such that the handoff decision is relatively resilient to short-term signal fading.

Note 2.— It is recommended that the algorithm include a signal quality threshold such that handoff can only take place when the signal quality is below the threshold. This recommendation is provided so handoff will not take place if the existing link is healthy and has ample margin.

Note 3.— It is also recommended that the handoff algorithm consider the Ground Subnetwork Address of the new ground stations in an attempt to minimize changes in the IDRP connection where possible by providing some preference to ground stations connected to the same router as the current connection.

5.3.1.2 Address identification

The link layer header shall contain the aircraft station radio address.

Note.— The ground station address is implied by the frequency and slot assignment.

5.3.1.2.1Address uniqueness. All aircraft stations radios shall be uniquely addressed by the 24-bit ICAO address in combination with the Radio Identifier indicated in the Address Type field.

5.3.1.2.2Broadcast addressing. The VDL shall support broadcast addresses that are recognized and acted upon by all appropriate receivers. Broadcast addressing service shall be as defined in Section 5.3.4.2.

5.3.2.2.2Address type. The address type subfield shall consist of three bits which indicate the usage of the remaining 24_-bits aircraft address. The most significant bit is bit 3 in Figure 5-4. Table 5-12 lists the address type options.

Note 1.— This field will indicate to the receiver whether the aircraft station address or broadcast address should be used to decode the FCS.

Note 2.— For the ground station, the MAC sub-layer can provide the 68-bit AircraftLocal Identifier via primitives to the DLS to determine which aircraft station address to use to extract the FCS.

5.3.2.3.4Toggle. The Toggle (T) subfield shall alternate between “0” and “1” for each unique acknowledged frame group (see 5.3.4.3.1.3) being sent. Reception of an ACK frame shall trigger the toggling of the bit. The receiver shall consider successive acknowledged frame groups with the same setting tof the T bit as duplicates and shall discard acknowledged frames whose T bit does not change value from the previously received frame group containing acknowledged frames. The receiver shall still send an ACK frame for each group correctly received. The DLS shall ignore this bit for unacknowledged frames.

Note.— The ground station needs to maintain separate T bit states for each aircraft connected.

- 80 –

5.3.4.3.1.3Frame grouping. Frame grouping, if supported, shall be performed at the DLS sub-layer using the MF bit of individual frames to indicate whether more frames are to follow or not to facilitate ungrouping at the receive end. Frame grouping shall observe the following rules:

a) The frame group shall not exceed the length indicated by the parameter N1, which is the maximum size of a frame group (including the frame headers of the individual frames).

b) Frames of the same priority, regardless of acknowledged frames and unacknowledged frames, shall be grouped first. Grouping of higher priority with lower priority frames shall be permitted, provided the addition of lower priority frames can be accommodated without requiring additional bursts.

c) With respect to frames requiring acknowledgement, only those with the same destination shall be grouped.

d)5.3.4.3.1.3.1The frame group shall be presented to the MAC sub-layer in a single reservation for media access. A frame group shall be treated as a single entity with respect to acknowledgement. Only the acknowledged frames within a frame group shall be retransmitted upon T1 timer expiration. The unacknowledged frames shall be dropped upon completion of the transmission and shall not be retransmitted.

Note.— The intention of allowing lower priority messages to be grouped is to allow otherwise wasted bandwidth to be used. This filling of excess space in a burst is not mandatory.

5.3.4.3.2 Reception acknowledgement

5.3.4.3.2.1Acknowledged frames. Discretely-addressed INFO and CTRL_RSP frames shall require acknowledgement of correct reception via the ACK frame. Upon correctly passing the FCS test of all frames requiring acknowledgement in a media access event (e.g. single frame or group), the DLS shall send a single ACK frame to the originator of the frame or group using the same radio identifier. If the FCS for any frame requiring acknowledgement in a media access event fails, all frames requiring acknowledgement (discretely-addressed INFO and CTRL_RSP frames) in the group shall be discarded.

Note.— The aircraft MAC sub-layer will remove ACK frames from the queue and send the M downlink (Acknowledge) burst instead.

5.3.4.3.2.2Unacknowledged frames. Broadcast INFO and all CTRL_CMD and ACK frames shall be unacknowledged. If any frame not requiring acknowledgement passes its FCS, then that frame should be processed.

5.3.4.3.3Transmission acknowledgement/retransmission. The DLS shall not send a new frame group to the MAC sub-layer until an acknowledgement frame has been received for the previous acknowledged frame group requiring acknowledgement or upon expiration of the T1 timer. When the T1 timer expires, the DLS shall provide the MAC sub-layer with the same frame group, minus any unacknowledged frames for retransmission.

Note.— The Media Access Control sub-layer is responsible for the determination of when to retransmit the message. The DLS must remain involved in the retransmission procedure as the MAC sub-layer cannot acknowledge correct receipt of a message until the DLS declares the message to have passed the FCS test.

5.3.5.1Aircraft DLS-MAC processing delay. The maximum processing delay allowable for the aircraft VDLaircraft radio to process an uplink ACK frame shall not exceed 200 ms under the worst-case aircraft loading conditions. The uplink ACK processing delay shall be defined as from the time that the aircraft MAC receives the uplink V/D (data) burst, which contains the uplink ACK, and ends with the uplink ACK being extracted by the aircraft DLS and forwarded to and received by the aircraft MAC to cause the T1 timer to be cancelled.

- 81 –

5.4.2.1.1Net Initialization. Before the link establishment procedure, the aircraft stationradio shall initialize to initial timing state 0 (TS0) and set parameters CTC1, CTC2 and CTC3 to max(f-50, t). The aircraft station radio shall enter TS0 either at the end of the transmission of the Leaving Net message (for discrete-addressed operation) or upon receipt of a new tuning command at the radio (for non-discrete-addressed operation). The first part of the link establishment procedure shall be the acquisition of the M uplink burst (LBAC 11 for 3T configuration, LBAC 5, 7 or 9 for 3S configuration, LBAC 4 or 6 for 2S1X configuration, LBAC 9 for 1V3D configuration, LBAC 7 for 1V2D configuration and LBAC 5 for other configurations) associated with the specified user group identifier. Upon receipt of two consecutive M uplink bursts containing the identical information (i.e. System Configuration, Squelch Window, Slot ID and Ground Station Code) in bits 5-8 of octet 2 and all of octet 3 of the uplink burst, the LME shall declare the net initialized and shall establish the net operating parameters based on the static information contained in bits 5-8 of octet 2 and all of octet 3. Configurations 3S and 2S1X shall be considered initialized when at least one of the wide area coverage slots provide two consecutive M uplink bursts containing the same information in bits 5-8 of octet 2 and all of octet 3. Upon net initialization, the aircraft station shall transition from TS0 to TS1.

5.4.2.1.1.1Upon receipt of two consecutive M uplink bursts containing the identicalconsistent information (i.ein the. System Configuration, Squelch Window, Slot ID and Ground Station Code) fieldsin octets 1-3 of the uplink burst, the LME shall declare the net initialized and shall establish the net operating parameters based on the static information contained in octets 1-3. In configurations 3S and 2S1X, the LME shall declare the net initialized when at least one of the wide area coverage slots provide two consecutive M uplink bursts containing the sameconsistent information in octets 1-3. Upon net initialization, the aircraft stationradio shall transition from TS0 to TS1.

5.4.2.1.1.2If the aircraft stationradio is unable to enter TS1, it shall search for ALTs to determine the number of slots in its net and enter TS2. If CTC3 equals parameter f before the aircraft stationradio enters TS2, it shall enter TS3. Prior to net initialization, voice access shall be inhibited.

5.4.2.1.1.3Transition from TS0 to TS2 or TS3 shall be prohibited (see Section 5.2.4.1.2) if an ALT received from a ground station (i.e., an uplink beacon from a ground station operating in another time slot) contains system configuration information that indicates that the aircraft radio has incorrectly tuned to a data slot.

Note 1.— Net Initialization is a prerequisite for the next step of Net Entry, however with Net Initialization only, basic voice services can be supported.

Note 2.— For a radio to initialize to TS2 directly, the rule requiring 2 consecutive M uplink bursts (with identical information) for initialization to TS1 does not apply. Only a single ALT is required for initialization to TS2. Initialization to TS2 is strictly controlled by the CTC1 and CTC2 counters in accordance with the rules specified in Tables 5-8b) and 5-8c).

5.4.2.1.2Net Entry. The net entry procedure shall begin with the aircraft station radio transmitting a Net Entry Request message following a successful net initialization. The transmission of the Net Entry Request message and the subsequent retransmission, if enabled, shall follow the procedures described in Appendix D. The Net Entry retransmission shall continue until a Net Entry Response message is received or the number of retransmissions reaches NL1. After NL1 retries (Table 5-10a)), the SN-SME shall be notified. If a Net Entry Response message is received, the aircraft station radio shall transmit a Poll Response message in the next MAC cycle. If the Net Entry Response message is of type “previous link preserved”, the aircraft TL3 timer (Table 5-10a)) shall be stopped.

Note.— All configurations that offer one or more data circuit(s) must support net entry (3V1D, 2V2D, 2V1D, 1V3D, 1V2D, and 3T). The voice only configurations (3V, 4V, 3S and 2S1X) may support net entry if discrete addressing is enabled by the ground user.

- 82 –

5.4.2.1.2.1In cases where multiple aircraft radios are installed on a single aircraft, the transmission of the Net Entry Request message and the initial Poll Response message from each radio shall be delayed by a different number of symbol periods.

Note.— This requirement is to prevent multiple messages in the same time slot from adding constructively, creating a situation where multiple radios have the same Local ID. A convenient way to implement this requirement would be to delay these particular down link transmissions by an amount equal to (SDI-1) symbol periods. The Source/Destination Indicator (SDI) is discussed in Appendix C.

5.4.2.1.2.21Upon receiving a Net Entry Request message, the ground shall check the aircraft ICAO address. If this address has been assigned a Local ID and the TL3 timer has not been started, the ground shall transmit the Net Entry Response message of type “previous link preserved.” When the Poll Response message is received by the ground station, the ground station shall respond by sending a Reservation Response message to the entering aircraft during the next MAC cycle. This message shall be either a reservation or a RACK if data slots were requested or a “no reservation” message if no data slots were requested.

5.4.2.1.2.2If the aircraft ICAO address has not been assigned a Local ID, the ground shall assign the next available Local User ID to the requesting aircraft. andThe ground station shall check to see if the aircraft ICAO address already has an active Local ID association within the same User Group. If so, the ground shall transmit the Net Entry Response message of type "previous link preserved." If not, the pending disconnect link table shall be examined. If the aircraft ICAO address is found in the table, the link status shall be reinstated, the TL3 timer shall be stopped, and the link port address shall be changed to the new radio station port address. The ground shall transmit the Net Entry Response message of type “previous link preserved”. When the Poll Response message is received by the ground station, the ground station shall respond by sending a Reservation Response, contained within a Supported Options message, to the entering aircraft during the next MAC cycle. This message shall be either a reservation or a RACK if data slots were requested or a “no reservation” indication if no data slots were requested.

5.4.2.1.2.3If the aircraft ICAO address is not found in the table and a Local ID has been assigned, the ground shall transmit the Net Entry Response message of type “no previous link”.

5.4.2.1.2.3If a ground station receives a duplicate Net Entry Request (i.e., from the same ICAO Address) prior to transmitting a Net Entry Response to the initial pending request, the duplicate shall be ignored.

5.4.2.1.2.4 When the Poll Response message is received by the ground station, the ground station shall respond by sending a Reservation Response message, contained within a Supported Options message, to the entering aircraft during the next MAC cycle. This message shall be either a reservation or a RACK if data slots were requested or a “no reservation” indication if no data slots were requested. If an aircraft receives a Net Entry Response message of type “no previous link”, it shall also form a join event message, as required by the supported network stack, and deliver it to the air router. If this is a new Local ID association for the ICAO address in the user group, the Supported Options message shall indicate Radio Identifier of 1. If there are already Local ID associations for the ICAO address in the user group, the Supported Options message shall indicate a Radio Identifier of the next available number. If all 3 Radio Identifiers are in use, the Radio Identifier field shall indicate 'Unavailable'. If the aircraft radio receives a Radio Identifier of 'Unavailable', it shall consider data service is not currently available for that radio on the net.

5.4.2.1.2.45If there are already 60 aircraftthe maximum allowable Local Identifiers assigned in the net when the Net Entry Request message is received, the ground LME shall respond with a Net Entry Response message with the aircraftlocal identifier field set to 0 indicating that the net is currently full. The aircraft radio shall then use an Aircraft ID of 61 when communicating with the ground station and other aircraft in the headers of the V/D (voice) bursts to indicate that it has not entered the net. An aircraft radio in this state shall not transmit downlink M bursts or V/D (data) bursts. In particular, it shall not reply to the Net Entry Response message. When an aircraft radio is in this state, it shall continue to monitor all Net Entry Response messages in order to detect a delayed Net Entry Response message addressed to it via its aircraft ICAO address. Upon receipt of such a message, the aircraft radio shall change its Local ID to the value provided. The ground LME shall provide a delayed Local ID to an aircraft radio request only if 1 instance of the ICAO address has been identified by the ground LME as being in delayed net entry in that net.

- 83 –

Note 1.— There is no limit on how many aircraft can use the aircraft ID of 61 simultaneously.

Note 2.— Aircraft radios receiving a “Full Net” indication in the Net Entry Response message cannot avail themselves of services requiring discrete addressing until a valid Local ID is provided via the delayed net entry process. From that point forward the aircraft radio can participate in all discrete-addressed services it can support. For example, data-capable radios can begin link negotiation via XIDs at this time.

Note 3.— In order to provide a delayed net entry capability, the ground station must maintain a time ordered list of potential new and existing ICAO address net entrants. As addresses become available, Local ID numbers should be distributed to waiting new-ICAO-address aircraft radios on a first-come-first-served basis.

5.4.2.1.2.56If the net has only NL5 or less free identifiers, the ground station shall provide an aircraft ID of 0 (indicating that the aircraft should use Aircraft ID = 61) to a new request by a radio whose ICAO address has already been assigned a local identifier that is still being polled. If NL6 does not equal -1 and when the ground station comes within NL6 identifier(s) of a full net and there are multiple Local IDs assigned to a particular aircraft ICAO address, the ground station shall send a Terminate Net message to free the Local ID from the most recent association that meets all of the following criteria: 1. No outstanding Urgent Downlink Request for the Local ID, 2. Local ID has not transmitted on the voice channel, 3. Local ID not associated with an active data link.The ground station should not reassign a Local ID for which it has sent a Terminate Net message until sufficient time has passed that the previous aircraft radio would time out the identifier due to lack of polling.If a duplicate Net Entry Request message is received it shall be discarded.

Note 1.— For further definition of NL5 and NL6, see Sections 5.4.3.2.6 and 5.4.3.2.7, respectively.

Note 2.— A Net Entry Request message is considered to be a duplicate if a ground station receives a retransmitted Net Entry Request while it is still processing a previously received Net Entry Request from the same aircraft. The above requirements provide the system with the ability to recover Local Identifiers utilized for multiple radio installations to support new users attempting to enter an almost full net. The capability is defined in a flexible manner as the acceptability of removing addresses previously assigned has not been determined.

5.4.2.1.2.67If the requesting aircraft does not receive a Reservation message from the ground in the next MAC cycle after transmitting the Poll Response message, it shall restart the net entry procedure.

Note.— The GNI provides a connection management function across multiple ground stations such that the same subnetwork connections are maintained with aircraft served by those ground stations. A description of the GNI function is provided in the Implementation Aspects of this manual.

5.4.2.1.2.78Supported Options. The final uplink in the net entry process includes the Supported Options message. The aircraft LME shall enable or disable the options indicated in the message. The options shall be as defined in Table B-9. The aircraft DLS shall utilize the appropriate Address Type fields consistent with the Radio Identifier indicated in the Supported Options message.

5.4.2.1.3.5Parameter Longevity. The parameters negotiated during net entry shall be applicable for the duration of the link (or until renegotiated)

5.4.2.1.3.5.1Net Entry Parameter Applicability. Parameters WE and RE apply to the next Net Entry attempt(s).

- 84 –

5.4.2.1.3.5.2Previous Link Preserved. The parameter set from the last link shall continue to apply if the ground station indicates ‘Previous Link Preserved’ during net entry and the TL3 timer has not yet expired.

5.4.2.3.2.1.4Aircraft initiation. The aircraft LME shall request a modification of parameters of a link by sending a CTRL_CMD_LPM to the ground LME containing the parameters per Table 5-15b). Optional and changed parameters shall be included if a non-default value is desired.

5.4.2.3.2.1.5General ground response. If the ground receives the CTRL_CMD_LPM frame, it shall confirm the link configuration by sending a CTRL_RSP_LPM frame containing the parameters from the CTRL_CMD_LPM that are accepted as well as any that are modified from those included in the initial request.

5.4.2.3.2.1.6Air Response. If the aircraft LME receives the CTRL_RSP_LPM, it shall confirm receipt via an acknowledgement M-burst.

5.4.2.3.2.2 Broadcast link parameter modification

5.4.2.3.2.2.1Ground initiation. The ground LME shall request a modification of parameters of all links by sending to all aircraft LMEs in the specified user group a broadcast-addressed CTRL_CMD_LPM to all aircraft LMEs in the specified user group containing the parameters per Table 5-15b).

5.4.2.3.2.2.2General Aircraft Response. The aircraft LME shall not respond to broadcast CTRL_CMD_LPM frames.

Note.— As there is no acknowledgement to broadcast CTRL frames, the ground station operator may transmit the frames multiple times to increase the likelihood of correct reception.

5.4.2.3.2.3Parameter longevity. The parameters negotiated shall be applicable for the duration of the link (or until renegotiated).

5.4.2.3.2.3.1Net Entry parameter applicability. Parameters WE and RE apply to the next Net Entry attempt(s).

5.4.2.5.3Polling system configurations. All aircraft shall be polled and respond with a Net Entry Request message per 2.9 of Appendix B to reconstruct associations between ICAO address and local ID. Polling will start at the highest aircraft ID 60for the system configuration and decrement by one each MAC cycle.

5.4.2.5.4Data-capable system configurations. If the radio is operating in a system configuration supporting data service, upon receiving the uplink Reservation Response field of the Recovery message indicating that it is granted a slot to transmit, the aircraft radio shall downlink a CTRL_CMD frame containing an Expedited Recovery XID, per 5.4.3.3.4.2.1, and a Network Initialization XID, per 5.4.3.3.4.1.3 and use of an ICAO Address of OxFFFFFF in place of the Aircraft Station Address of the DLS Header, per 5.3.2.2.1, so that the ground station may easily extract the FCS. The Expedited Recovery XID indicates the actual aircraft addressing information as well as information needed to recover the data scheduler. The Network Initialization XID indicates the desired subnetwork connection. The aircraft shall only transmit those network information and network subparameters needed to initialize the network. The CTRL_CMD frame downlinked shall be limited to a single burst transmission so that the ground system can deterministically schedule access during recovery. Additional parameters and network connections should be negotiated after the recovery process is complete, with the aircraft initiating the Link Maintenance procedures of Section 5.4.2.3.2.

Note.— For system configurations 3V1D and 2V1D, all local IDs can be polled faster than all aircraft can access the data subchannel. For 2V2D and 3T, this will also help expedite the procedure, as well as simplify the

- 85 –

implementation. As such, it may be desired to allow the ground station to skip sending an uplink Recovery message to a local ID that is known to not be assigned due to lack of response to the poll. It may also be desirable to start at the high end of the local identifier range (i.e. aircraft ID 60 or 120, as applicable) and count down to prevent overlap with the data reservations for recovery information so as to maximize its ability to discover open local identifiers before a data slot is wasted. Once the polling reaches the point where the local ID has already been queried via the data reservation, the polling can ceaseskip local IDs that have responded to the uplink Recovery message to determine if there are any enhanced-voice-only users.

5.4.2.5.7Verification. At the completion of the recovery procedure for data-capable system configurations, the ground station shall broadcast a Connection Check XID within a CTRL_CMD_LPM message containing the correlations of local identifier with ICAO aircraft station address and radio identifier for all known aircraft in the user group. Aircraft should verify their entry for correctness. If their aircraft station address is not contained within the message, the aircraft should attempt Link Establishment per 5.4.2.1.

5.4.2.5.81V3D/1V2D System Configuration. For system configurations 1V3D and 1V2D performing expedited recovery, the ground station shall uplink 2 reservation response messages in the M uplink Recovery message consistent with Section 5.4.2.5.2 indicating the use of the B-slot for the 2 TDMA frames covered by the MAC cycle. The aircraft whose local ID is 1 greater than the aircraft identified in the reservation response message shall assume a reservation response granting them access to slot C in the same TDMA frame.

5.4.2.5.8.1For system configuration 1V3D, an aircraft whose local ID is 2 greater than the aircraft identified in the reservation response message shall assume a reservation response granting them access to slot D in the same TDMA frame.

5.4.2.6.1.2Upon receipt of aAfter the Leaving Net message is received, the ground shall first verify that the last 9 bits of the decoded message are all zero. If any of the bits are not zero, the entire message shall be discarded. If the message is verified, the ground shall start the TL3 timer, change link status to pending, stop polling the aircraft, and make the Local User ID assigned to the aircraft radio available for reassignment. When the TL3 timer expires, the link shall be changed to disconnect status, and a leave event message, as required by the supported network stack, shall be formed and delivered to the ground router(s).

5.4.2.7.3Same GNI. When an aircraft transfers to a new net which is served by the same GNI, the data link between the GNI and the aircraft shall be redirected from the old ground station to the new ground station. To accomplish this, first, the aircraft shall execute the explicit link release procedure described in Section 5.4.2.6.1 while in the old net. Then, the link establishment procedure described in Section 5.4.2.1 shall be used to reinstate the link status when the aircraft enters into the new net. The subnetwork virtual connection, if applicable, shall remain undisturbed, while the link layer switches nets.

5.4.2.7.4Different GNI. When an aircraft transfers to a new net which is served by a different GNI, the data link between the old GNI and the aircraft shall be disconnected as described in Section 5.4.2.6.1 or Section 5.4.2.6.2. A new data link shall be established following the link establishment procedure described in Section 5.4.2.1.

5.4.3.2.1Parameter NL1 (maximum net entry request retry. The Net Entry Request message shall be retransmitted no more than NL1 times. When the number of Net Entry Request retransmissions exceeds NL1 before a Net Entry Response message is received, subnetwork layer system management entity (i.e. SN-SME) shall be informed.RESERVED

- 86 –

5.4.3.2.3Timer TL3 (disconnect delay timer). Timer TL3 defines the time interval that the LME shall wait before the link is changed to disconnect state, and a Leave_Event message, as required by the supported network stack, is sent to the local router.

Note.— This timer is uUsed by the LME to support link redirection across multiple ground stations.

5.4.3.2.4Timer TL4 (polling interval). Timer TL4 shall represent the maximum number of elapsed MAC cycles between receipt of two consecutive polls. Upon receipt of a Poll Request message, the aircraft LME shall reset the TL4 timer. Timer TL4 shall be started upon receipt of a poll. Upon expiration of this timer, the LME shall notify the DLS, SNDCF, and the Voice Unit and attempt to re-establish the link by sending a Net Entry message to the ground station.

5.4.3.2.5Parameter WE (net entry retransmission delay). The parameter WE specifies the number of MAC cycles that an aircraft radio shall wait before attempting a Net Entry Request message retransmission.

5.4.3.2.6Parameter NL5 (multiple radio local identifier buffer). The parameter NL5 specifies the number of free local IDs at which point the ground station LME shall stop providing addresses to ICAO addresses attempting net entry that already have at least one entry in the table for that net, per Section 5.4.2.1.2.

5.4.3.2.7Parameter NL6 (free address recovery). The parameter NL6 specifies the point whereby the ground station LME shall begin recovering local IDs from aircraft which have multiple radios participating in discrete-addressed services in the same net consistent with Section 5.4.2.1.2. A value of -1 shall indicate that the ground LME shall not recover addresses.

5.4.3.3.1.1General public parameters. The general public parameters’ XID information field shall be encoded as defined in ISO 8885, with the addition of private parameter data link layer subfields as defined in ISO 8885. The format identifier, hexadecimal 82, shall be used (per ISO 4335.2, Annex C) to identify the public parameter list identified in ISO 8885. The VDL Mode 3 shall use the public parameter group ID of hexadecimal 80 to indicate negotiation of the common parameters. The public parameter set ID shall be included in CTRL frames if other public parameters are included; the public parameter set ID shall not be included in the CTRL frame if other public parameters are not included. Only the public parameters Unique Identifier (also known as Parameter Set Identifier), Maximum information field length - transmit and receive (N1), Acknowledgment timer (T1), and Retransmission attempts (N2) shall apply to the VDL Mode 3 system.

Note.— A Control frame example is contained in Part I of theis Manual on the Implementation of the Very High Frequency (VHF) Digital Link Mode 3 (VDL Mode 3) of ICAO.

5.4.3.3.3Mode 2 private parameters. The following Mode 2 parameters shall also be available for use by the Mode 3 SystemUse of any of the parameters defined for VDL Mode 2 use requires the inclusion of the Parameter Set Identification parameter with a value of ‘V’, as defined in Section 5.4.3.3.3.1.1.

5.4.3.3.4.1.3Network initialization parameter. The network initialization parameter defines the network support which the avionics VDL Mode 3 installation can perform. It is used to pass any necessary information to initialize the subnetwork protocol or any compression. Table 5-27a) defines the format of this parameter set. The parameter length field shall contain the number of networks present. The Network Type shall be encoded per Table 5-27b). For each network, a network length field shall indicate the number of subparameters. If no subparameters are needed for a network, then that network’s length field shall indicate zero.

- 87 –

5.4.3.3.4.1.3.1 Note 1.— As many networks as applicable can appear, as defined in Table 5-27b).

Note 2.— It is assumed that the ground system will support all options and the aircraft system will support at least one.

A Network Type value of zero indicates that the VDL Mode 3 system is carrying ISO 8208 packets which shall be transferred to an ATN ISO 8208 data communication equipment (DCE) at the remote end. The ISO 8208 packets shall be compressed using the VDL Mode 3 PLP Compression as defined in 6.2.2.1. A Network Type value of one indicates that the VDL Mode 3 system is carrying ATN CLNP packets which shall be transferred to an ATN-conformant CLNP service at the remote end. The CLNP packets shall be compressed using the VDL Mode 3 compression as defined in 6.2.2.2. A Network Type value of two indicates that: (1) The VDL Mode 3 system is carrying ISO 8208 packets which shall be transferred to an ATN ISO 8208 DCE at the remote end; and (2) No compression is performed. Similarly, a Network Type value of three indicates that the VDL Mode 3 system is carrying ATN CLNP packets which shall be transferred to an ATN-conformant CLNP service at the remote end with no compression. A Network Type Value of four indicates that no (sub)network protocol shall be used. Data shall be sent to the remote end unmodified with no (sub)network operands. A Network Type Value of five indicates that the VDL Mode 3 system is carrying ATN frame mode packets which shall be transferred to an ATN frame Mode SNDCF at the remote end without compression within the subnetwork. Other Network Type values are reserved and are not supported.

Note 31.— The ATN designation tells the VDL Mode 3 system to support any ATN-specific requirements. For example, ATN join and leave events will be generated when network types of 0, 1, 2, 3 or 5 are sent from the aircraft to the ground.

Note 4.— An indication of ‘No Compression’ in this parameter only indicates that the VDL Mode 3 subnetwork performs no compression. This does not indicate or prohibit any compression by the network or higher layers.

Note 5.— The ground station will support all Network Types.

Note 26.— The use of the raw interface requires that the largest packet received with the addition of the payload and DLS frame header be a maximum of 930 bytes as no segmentation of the data is possible.

5.4.3.3.4.2Air-initiated information private parameters. An aircraft LME shall use aircraft-initiated information private parameters to inform the ground about the aircraft’s Mode 3 capabilities or desires. Ground LMEs shall not send these parameters, except to confirm the Former GNIp.

5.4.3.3.4.2.1Expedited recovery parameter. The Expedited Recovery parameter shall be used by the aircraft to communicate its MAC sub-layer state to allow the ground MAC sub-layer to reload its state information. Table 5-29 defines the format of this parameter. The aircraft address shall be the 24-bit ICAO aircraft station address. The Reservation Status field shall include a 2-bit priority field, and a 4-bit Number of Slots field indicating at what priority and how many slots are needed for any outstanding data requests for the aircraft , and a 2-bit Radio Identifier encoded as Table B-10. Upon reception by the ground station, this information shall be forwarded to the MAC sub-layer.

5.4.3.3.4.3Ground-initiated modification private parameters. A ground LME shall use the ground-initiated modification parameters to change the value of various Mode 3 parameters in one or more aircraft. Aircraft LMEs shall not send an XID with these parameters, except to confirm a change.

5.4.3.3.4.3.1NM1 (maximum retry) parameter. The value of NM1 (used by aircraft) shall be encoded as an unsigned 8-bit integer per Table 5-32.

- 88 –

5.4.3.3.4.3.2RR (reservation request randomizer) parameter. The value of RR (used by aircraft) shall be encoded as an unsigned 8-bit integer per Table 5-33. This negotiation shall only affect the parameter for the current system configuration.

5.4.3.3.4.3.3WR (reservation request delay) parameter. The value of WR (counted in MAC cycles and used by aircraft) shall be encoded as an unsigned 8-bit integer per Table 5-34. This negotiation shall only affect the parameter for the current system configuration.

5.4.3.3.4.3.4RE (net entry randomizer) parameter. The value of RE (used by aircraft) shall be encoded as an unsigned 8-bit integer per Table 5-35. This negotiation shall only affect the parameter for the current system configuration.

5.4.3.3.4.3.5RL (leaving net randomizer) parameter. The value of RL (used by aircraft) shall be encoded as an unsigned 8-bit integer per Table 5-36. This negotiation shall only affect the parameter for the current system configuration.

5.4.3.3.4.3.6Timer T3 (link initialization time) parameter. The value of timer T3 (in MAC cycles used by aircraft) shall be encoded as an unsigned 8-bit integer per Table 5-37.

5.4.3.3.4.3.7NL1 (maximum net entry request retry) parameter. The value of parameter NL1 (used by aircraft) shall be encoded as an unsigned 8-bit integer per Table 5-38.RESERVED

5.4.3.3.4.3.8Timer TL3 (disconnect delay timer) parameter. The value of timer TL3 (in seconds used by aircraft) shall be encoded as an unsigned 8-bit integer per Table 5-39.

5.4.3.3.4.3.9Timer TL4 (polling interval) parameter. The value of timer TL4 (in seconds used by aircraft) shall be encoded as an unsigned 16-bit integer per Table 5-40.

5.4.3.3.4.3.10WE (net entry request delay) parameter. The value of WE (in MAC cycles used by aircraft) shall be encoded as an unsigned 8-bit integer per Table 5-41. This negotiation shall only affect the parameter for the current system configuration.

5.4.3.3.5.2NL5 (multiple radio local identifier buffer) parameter. The NL5 counter, used by the ground station, shall be encoded as an unsigned 8-bit integer per Table 5-46.

5.4.3.3.5.3NL6 (free address recovery) parameter. The NL6 counter, used by the ground station, shall be encoded as a 1-bit sign (0=positive, 1=negative), s, and an unsigned 6-bit integer per Table 5-47.

6.2.5 Subnetwork interface support

6.2.5.1 Aircraft subnetwork interface support

Aircraft radios support data link services shall provide the raw interface for verification purposes.

6.2.5.1.1Aircraft radios supporting point-to-point data link services shall also provide at least one of the subnetwork interfaces supported by the ground.

6.2.5.2 Ground station subnetwork interface support

- 89 –

The ground station shall support the ATN/ISO8208/VDL Mode 3 PLP Compression and ATN/ISO8473/ISO9542/VDL Mode 3 CLNP Compression subnetwork interfaces.

6.2.6 Raw subnetwork interface operation

6.2.6.1 Protocol identifier

The raw subnetwork interface shall identify the contents of the user data via a signature component identifier consistent with an ISO/IEC 9577 header. ISO 9577 Initial Protocol Identifier (IPI) byte encoding of all-ones (0xFF) is reserved by ISO/IEC TR 9577 to allow for the specification of additional subnetwork protocols beyond those explicitly defined in the ISO document. When this form of IPI encoding is employed, a second octet (the Extended Initial Protocol Identifier (EIPI)) is used to specify the actual subnetwork protocol in use. The first one or two octets following the Payload byte shall indicate the protocol being passed via this interface with the user data to follow. The encoding of the IPI and EIPI bytes shall be in accordance with Table 6-2.

6.2.6.2 Supported Protocols

6.2.6.2.1ACARS over VDL3. The VDL Mode 3 raw subnetwork interface may be used to pass ACARS data. After the IPI and EIPI bytes, the ARINC 618/620 message shall follow.

6.2.6.2.2Flight Information Service - Broadcast. The VDL Mode 3 raw subnetwork interface may be used to pass FIS-B messages in a broadcast uplink V/D (data) burst. After the IPI and EIPI bytes, the FIS-B message shall follow.

6.2.6.2.3Reserved for Verification. The combination of an IPI of 0xFF and EIPI of 0 are reserved for use in MOPS testing.

7. VDL MODE 3 MOBILESUBNETWORK DEPENDENT

CONVERGENCE FUNCTION (SNDCF)

7.1 Introduction

The VDL Mode 3 mobile shall support the raw interface and one or more of the other defined SNDCFs. The first type of SNDCF is ISO 8208 based and corresponds to that described in ICAO Doc 9705 for mobile SNDCF. Procedures for this SNDCF are defined in 7.2. The second type of SNDCF, denoted frame-based SNDCF, shall uses the procedures defined in 7.3. The third type of SNDCF, denoted ATN Frame Mode SNDCF, is defined in Section 7.4

Note.— The SNDCF is termed frame-based because it uses the VDL Mode 3 frames without the need for an additional protocol (viz. ISO 8208 SNDCF) to transfer network packets. VDL Mode 3 achieves independence from the network protocol by identifying the payload of each frame. Upon receipt of a frame, the payload field is examined and control is passed to the protocol identified.

7.3.2.3 Disposition of CLNP AND and ISH packets

When ATN connectivity has been specified during the net entry procedures, all CLNP and ISH network packets received by the SNDCF from its remote peer(s) shall be transferred to the local ATN router.

- 90 –

7.4 The ATN frame mode SNDCF

The VDL Mode 3 shall support the ATN Frame Mode SNDCF for mobile subnetworks as defined in the ICAO Doc 9705, Sub-Volume V, 5.7.8.

8.2.2 Voice burst framing

Exactly six voice frames shall be transmitted in a single V/D (voice) burst during full rate operations (TS1 and TS3). Exactly five voice frames shall be transmitted in a single V/D (voice) burst during truncated voice mode operations (TS2).

8.3.1.3If the net entry procedure has not been executed, the value of Group ID shall be the same as the value of Slot Beacon ID upon used for net initialization.

8.4.1 Link establishment

8.4.1.1Link initialization. Upon Net Initialization (see Section 5.4.2.1.1), the aircraft LME shall provide the Voice Unit with the net parameters as part of the LME’s net initialization procedure. These parameters shall be the Slot Beacon ID for transmission, possible additional Beacon IDs for reception and Squelch window.

Note.— For net initialization directly from TS0 to TS3, there will be no received Beacon Slot ID information and the Squelch window will be wide open. Transmitted headers should used the Group ID used for the attempted net initialization.

8.4.2.1Transmit delay. The delay between speech input and output of the associated V/D (voice) burst shall be less than or equal to 175 ms. This delay is measured from the input of the audio signal to the radio to the beginning of the transmission of the user data for the first vocoder frame in the TDMA frame.

Note.— The delay budget includes up to 35 ms additional delay for ground network functions, such as modem buffering and ground telecommunications delay. It is recognized that some ground implementations may exceed this; however, studies have indicated that an additional 80 ms of delay can be incurred without significant impact to air traffic control operations.

8.4.3.2.1Truncated mode encoding. The Voice Unit shall reduce the rate of the clocking signal to the vocoder to 5/6 of it normal rate, thereby reducing the output data to a 4 000 bps rate.

Note 1.— This has the effect of increasing the voice frame size to 24ms for truncated operation and only generating five vocoder frames per TDMA frame

- 91 –

. Note 2.— The DVSI VC-20 ATC-10B and bit-exact fixed point implementations do not include truncated mode operation as provided from DVSI. Implementators will be responsible to add the truncated mode capability to this vocoder subsystem.

8.4.3.2.43.1The MID field of the V/D (voice) burst header shall be set to Downlink voice, TS2, to indicate to the receiver that the burst is truncated.

8.4.3.2.53.2If the Voice Unit is notified that it is to operate in normal or free running mode, the voice unit shall transmit full length V/D (voice) bursts. In normal mode, the MID field of the V/D (voice) burst header shall be set to Uplink voice for the ground station and to Downlink voice, TS1, for the aircraft stationradio. In free running mode, the MID field of the V/D (voice) burst header shall be set to Downlink voice, TS3. The Voice Unit shall transmit the bursts at the nominal CTS signal.

8.4.3.54Urgent Downlink Request (UDR). When the CTS indication is cleared and uUpon user command, the MAC sublayer shall formulateuse a random access Reservation Request M burst message (or a Poll Response M burst message, (as appropriate) to transmit an Urgent Downlink Request by setting the Voice Request field (Appendix B, Section 2.10.4) to “Request Priority Access” and shall transmit the M burst message at the next available opportunity. The slots used for sending and retransmitting the urgent/priority message are chosen using the same rules that are used for making a data reservation request. The aircraft station will continue sending the message until it receives a RACK. The rules for transmitting UDR bursts are the same as the rules specified for transmitting a data reservation request.

8.4.3.4.1UDR Cancellation. Note: Sending of a Reservation Request or Poll Response with the Voice Signal field indicating ‘No Request’ shall be is used to cancel pending Urgent Downlink Requests.

8.4.3.4.2UDR Pending. The aircraft radio shall provide indications to the user whether a UDR is pending.

Note 1.— “UDR Pending” indicates that a UDR has been initiated by the radio.

Note 2.— How UDR indications are provided to the user is an implementation issue and is outside the scope of these provisions.

8.4.3.5.14.3Urgent Downlink Request Not Supported. If the Supported Options message or acknowledging RACK indicates Urgent Downlink Request is not supported, the aircraft radio shall provide an indicator for the user that the request has failed. Further attempts to transmit Urgent Downlink Request shall be inhibited while the aircraft remains in that net.

8.4.3.45Anti-blocking. If an aircraft radio has its PTT keyed and CTS is cleared (indicating the channel is busy), voice transmission shall be inhibited until PTT is reasserted by the user (after which the voice access rules of Table 5-9 need to be reapplied). Upon user command, the aircraft station shall unblock the cleared CTS flag when the cleared flag is due to the channel being occupied by TS2 or TS3 voice activity.

Note 1.— The purpose of this requirement is to prevent a second user from transmitting a partial communication immediately following the end of transmission of another user, which could lead listeners to misunderstand the communications.

8.4.3.5.1 The aircraft radio shall be commandable to allow transmission when the channel is being occupied by TS2 or TS3 voice activity.

Note 2.— As for the ability to disable the transmission inhibit, Table 5-9 is still applicable to determine whether the transmitter is allowed to output to the voice channel.

- 92 –

8.4.4.1.2Vocoder processing. A received V/D (voice) burst with MID field of Uplink voice, or Downlink voice of: TS1, or Downlink voice: TS3 shall be processed at the normal vocoder rate of 4 800 bps. A received V/D (voice) burst with MID field of Downlink voice,: TS2, shall be processed at the truncated rate of 4 000 bps.

8.4.4.2.1If bursts are missing, the Voice Unit shall not generate clicks and pops during transition from presence of burst to absence of burst, in the absence of bursts, and transition from absence of burst to presence of burst. If bursts are missing, the voice unit shall slowly attenuate the speech to silence (or pseudo-noise) . The Voice Unit shall resume normal operation when receipt of the V/D (voice) bursts continues.

TABLES FOR THE MANUAL ON VHF DIGITAL LINK (VDL) MODE 3TECHNICAL SPECIFICATIONS

Table 5-1. LBAC timing offset values for configurations4V, 2V2D, 3V, 3V1D and 2V1D

System Config.

User Group SlotBeacon ID

Media Access Timing for each LBAC (See 5.2.2.5.1) relative to timing reference point (Symbol periods)

Even frame Odd frame

1 2 3 4 5 6 7 8

4 slots per frame (normal range)

4V A-D -1260 -1195 N/A N/A 0 65 N/A N/A

3V1D A -1260 -1195 -315 -250 0 65 945 1010

B -1260 -1195 -630 -565 0 65 630 695

C -1260 -1195 -945 -880 0 65 315 380

2V2D A-BD -1260 -1195 -630 -565 0 65 630 695

3 slots per frame (long range)

3V A-C -1260 -1142 N/A N/A 0 118 N/A N/A

2V1D A/C -1260 -1142 -420 -302 0 118 840 958

B/D -1260 -1142 -840 -722 0 118 420 538

Table 5-23a). LBAC description and timing for 3S configurations

LBAC # Applicability Description Media Access Timing for each LBAC relative to timing reference point

(symbol periods)

1 air only M downlink poll response or RA -1260

2 air, ground V/D (voice) even frame -1142

3 ground only V/D (voice) even frame -722


5 ground only M uplink and timing reference point 0

6 air, ground V/D (voice) odd frame 118


8 ground only V/D (voice) odd frame 538



Table 5-23b). LBAC description and timing for 2S1X portion of configuration

LBAC # Applicability Description Media Access Timing for each LBAC relative to timing reference point

(symbol periods)

1 air only M downlink poll response or RA -1260

2 air, ground V/D (voice) even frame -1142



5 air, ground V/D (voice) odd frame 118



Table 5- 2c 3d : LBAC Description and Timing for 1V3D Configuration

LBAC # Applicability Description Media Access Timing for each LBAC relative to timing

reference point (symbol periods)

1 Air only M downlink poll response -12602 Air/Ground V/D (voice) even frame -11953 Air only M downlink ACK or RA -9454 Air/Ground V/D (data) even frame -8805 Air only M downlink ACK or RA -6306 Air/Ground V/D (data) even frame -5657 Air only M downlink ACK or RA -3158 Air/Ground V/D (data) even frame -2509 Ground only M uplink, Timing Reference Point 010 Air/Ground V/D (voice) odd frame 6511 Air only M downlink ACK or RA 31512 Air/Ground V/D (data) odd frame 38013 Air only M downlink ACK or RA 63014 Air/Ground V/D (data) odd frame 69515 Air only M downlink ACK or RA 94516 Air/Ground V/D (data) odd frame 1010

Table 5- 2d 3e : LBAC Description and Timing for 1V2D Configuration

LBAC # Applicability Description Media Access Timing for each LBAC relative to timing reference

point (symbol periods)

1 Air only M downlink poll response -12602 Air/Ground V/D (voice) even frame -11423 Air only M downlink ACK or RA -8404 Air/Ground V/D (data) even frame -7225 Air only M downlink ACK or RA -4206 Air/Ground V/D (data) even frame -3027 Ground only M uplink, Timing Reference Point 08 Air/Ground V/D (voice) even frame 1189 Air only M downlink ACK or RA 42010 Air/Ground V/D (data) odd frame 53811 Air only M downlink ACK or RA 84012 Air/Ground V/D (data) odd frame 958

Table 5-3c). Relative timing relationship for TRPsof different user groups for all system configurations

Relative timings among TRPs of different user groups (symbol periods)

User GroupBeacon ID

4V 3V1D 2V2D1 3T*2 3V 3S**3 2S1X***3 2V1D1 1V3D ****4

1V2D ****4

A 0 0 0 N/A 0 0 0 0 0 0

B 315 315 315 0 420 420 N/A 420 0 0

C 630 630 N/A0 0 840 840 840 N/A0 0 0

D 945 N/A N/A315 0 N/A N/A N/A N/A420 0 0

1 For 2V2D/2V1D configurations, Group IDs A and C are treated as one user group and Group IDs B and D as another user group.

*2 For 3T configuration, Group IDs B, C and D are treated as one user group.**3 For 3S configuration, Group IDs A, B and C are treated as one user group.

*** For /2S1X configurations, Group ID’s A and C are treated as one user groupis implied regardless of the Beacon ID indication.

**** 4 For 1V3D and 1V2D configurations, Group IDs A, B, C,and D are treated as one user group.

Table 5-4. System configurations and Addressing Schemes

SystemConfig

Services to Each User

Group

Valid Channel

Label Suffix*(User

Group)

Valid Beacon ID**

Local User ID*** Poll Slot/ Poll

Response Slot

GroupID

Aircraft IDAircraft

w/ Discrete Address

Aircraft w/No

Discrete Address

GndStation

Normal Range

(4-Slot)

4V Dedicated voice circuit

ABCD

ABCD

ABCD

1-601-601-601-60

61616161

0000

A/AB/BC/CD/D

3V1D Dedicated voice circuit

w/shared data circuit

ABC

ABC

ABC

1-601-601-60

616161

000

A/AB/BC/C

2V2D Dedicated voice circuit w/dedicated data circuit

A

B

A or C

B or D

ACBD

1-601-601-601-60

61-

61-

0-0-

A/A

B/B


A A, B, C or D ABCD

1-601-601-601-60

61---

0---

A/AA/AA/AA/A

3T Demand Assigned voice and

data

A A BCD

1-601-601-60

61--

0--

A/A

Long Range

(3-Slot)

3V Dedicated voice circuit

ABC

ABC

ABC

1-601-601-60

616161

000

A/AB/BC/C

2V1D Dedicated voice circuit

w/shared data circuit

A

B

A or C

B or D

ACBD

1-601-601-601-60

61-

61-

0-0-

A/A

B/B


A A, B, C or D ABCD

1-601-601-601-60

61---

0---

A/AA/AA/AA/A

3S Dedicated voice circuit

with 3 station

diversity

A A, B or C A 1-60 61 0 A, B & C/A

2S1X Dedicated voice circuit

with 2 station

diversity

A A or C A 1-60 61 0 A & C/A

* “Channel Label Suffix” corresponds to the last digit of the VDL Mode 3 channel label defined in Annex 10, vol V, Chapter 4.

** The term “Valid Beacon ID” is defined as the first 2 bits of the third octet in the uplink M burst.*** Local User IDs not specifically identified in the Table are spares.

System Config. User groups supported/identifying time slots*

Services to each group

V/D slots assigned to each group

Normal Range 4V 4/(A, B, C, D) Dedicated voice circuit A, B, C or D

3V1D 3/(A, B, C) Dedicated voice circuit w/shared data circuit

A/D, B/D or C/D

2V2D 2/(A, B) Dedicated voice circuit w/dedicated data circuit

A/C or B/D

3T 1/(A) Demand assigned voice and data

B/C/D

Long Range 3V 3/(A, B, C) Dedicated voice circuit A, B or C

3S 1/(A, B, C) Dedicated voice circuit with 3 station diversity

A/B/C

2S1X 1/(A, C) Dedicated voice circuit with 2 station diversity

A/C

2V1D 2/(A, B) Dedicated voice circuit w/shared data circuit

A/C or B/C

* The term “identifying time slots” is defined as the value in the Slot Number field in the Beacon2 word of the uplink M channel.

Table 5-6. Access rules for M downlink

Config-uration

LBAC # Utilization of LBAC

RA ACK4V, 2V2D, 3V, 3V1D and 2V1D

1 LBAC 5 in previous MAC cycle did not request a poll response

No data acknowledgement allowed

3 LBAC 4 in previous MAC cycle did not have an uplink V/D (data) burst Header segment with Message ID* = 100 (Uplink Data – Acknowledged) and EOM** = 1

LBAC 4 in previous MAC cycle had an uplink V/D (data) burst Header segment with Message ID = 100 (Uplink Data – Acknowledged) and EOM = 1

7 LBAC 8 in previous MAC cycle did not have an uplink V/D (data) burst Header segment with Message ID = 100 (Uplink Data – Acknowledged) and EOM = 1


3T 1 Unconditional RA No data acknowledgement allowed

















Config-uration


RA ACK



3S 1 LBAC 5, 7 or 9 in previous MAC cycle did not request a poll response


2S1X 1 LBAC 4 or 6 in previous MAC cycle did not request a poll response


1V3D 1 N/A (Poll Response only) No data acknowledgement allowed

3 LBAC 4 in previous MAC cycle did not have an uplink V/D (data) burst header segment with Message ID = 100 (Uplink Data – Acknowledged) and EOM = 1

LBAC 4 in previous MAC cycle did not have an uplink V/D (data) burst header segment with Message ID = 100 (Uplink Data – Acknowledged) and EOM = 1











1V2D 1 N/A (Poll Response only) No data acknowledgement allowed





Config-uration


RA ACK





* Defined in 3.1.1 of Appendix B. **Defined in 3.1.2 of Appendix B.

Table 5-7a). Aircraft MAC service system parameters

Symbol Parameter name Min. Max. Mode 3 Default Increment

t Truncation(MAC cycles)

1 256 50* 1

f Free Running(MAC cycles)

1 256 250* 1

NM1 Maximum Reservation

Request Retry

1 128 20 1

RE Net Entry Randomizer

1 256 ** 1

RR Reservation Request

Randomizer

1 256 ** 1

RL Leaving Net Randomizer

1 256 **0 1

WR Reservation Request Delay (MAC Cycle)

1 256 ** 1

* Based on the specified clock accuracy of 5 PPM for aircraft radios and 2 PPM for ground radios, plus 1 PPM timing shift due to a 1 Mach aircraft.

** Value depends upon system configurations and is defined in Table 5-7b).

Table 5-7b). Aircraft MAC service system parameters

System configuration

Symbol Parameter name 4V 3V1D 2V2D 3T 3V 2V1D 3S 2S1X 1V3D 1V2D

RE Net Entry Randomizer

(Note 2)

3 8 8 8 3 8 3 3 8 8

RR Reservation Request Randomizer

(Note 2)

3 8 8 8 3 8 3 3 8 8

RL Leaving Net Randomizer

3 8 8 8 3 8 3 3

WR Reservation Request Delay (MAC cycle)

1 2 2 2 1 2 1 1 2 2

Note 1.— RE, RR, RL and WR are system configuration dependent.Note 2.— Units for RE, and RR, RL are in number of available time slots.

Table 5-8b). Timing states for Non-3T configurations

Timing state CTC1 value CTC3 value

Initial TS0*** Max(f** -50, t*) max(f** -50, t*)

Normal TS1 0 ≤ CTC1< t* 0 ≤ CTC3 < t*

Truncate TS2 CTC1 ≥ t* CTC3 > f**

Free Running TS3 CTC1 ≥ t* CTC3 ≥ f**

* Defined in 5.2.3.1** Defined in 5.2.3.2*** Net Initialization state, special rules apply (see 5.4.2.1.1)

Table 5-8c). Timing states for 3T configurations

Timing state CTC1 value CTC3 value

Initial TS0*** Max(f** -50, t*) max(f** -50, t*)

Normal TS1 0 ≤ CTC1< t* 0 ≤ CTC3 < t*

Transition (Non-Transmit) CTC1 ≥ t* CTC3 < f**

Free Running TS3 CTC1 ≥ t* CTC3 ≥ f**

* Defined in 5.2.3.1** Defined in 5.2.3.2*** Net Initialization state, special rules apply (see 5.4.2.1.1)

Note.— The timing state controls operation of the Voice Unit (Section 8).

Table 5-9. Conditions for voice media access (Aircraft StationRadio)

Voice signal field* encoding

Timing state Ground access Occupied Voice channel idle

Initial TS0** No access No access No access

Normal TS1 No access Access permitted if access granted in previous MAC

cycle

Access***

Truncate TS2**(non-3T)

Access Access Access

Transition**(3T only) No access No access No access

Free Running TS3** Access Access Access

* Defined in Appendix B, 2.1.4.** The aircraft stationradio does not receive Voice Signal Field Encoding information in these cases.

*** See Section 5.2.4.2.4 for exception.

Table 5-10a). LME system parameters

Symbol Parameter name Minimum Maximum Mode 3 default Increment

WE Net Entry Request Delay (MAC cycle)

1 256 *See Table 5-10b)

1

NL1(air)

Maximum net entry request retry 1 128 20 1

NL2(gnd)

Polling reply counter 1 255 3 1

TL3 Disconnect delay timer (seconds)

5 30 10 1

TL4(air)

Polling interval(MAC_cycles)

1 600 See Table 5-10b) 1

NL5 (gnd)

Multiple radio local identifier buffer

1 240 5 1

NL6 (gnd)

Free address recovery counter -1 60 -1 1

*Table 5-10b). WE and TL4 parameter default values

System configuration

Symbol Parameter name 4V 3V1D 2V2D 3T 3V 2V1D 3S 2S1X 1V3D 1V2D

WE Net Entry Request Delay (MAC cycle)

1 2 2 2 1 2 1 1 4 4

TL4 (air)

Polling interval (MAC_cycles)

480 120 120 120 480 120 240 240 480 480

Note.— This table takes into account the possibility that the first six system configurations might be used in sectors with site diversity.

Table 5-11. MAC data frame transmission system parameters

Symbol Parameter name Minimum Maximum Mode 3 default Increment

T1

(MAC cycles)

Delay before retransmission (ground)

1 40 2 1

Delay before retransmission (aircraft)

2 40 10 1

T_ack

(MAC cycles)

Maximum delay allowable

for ACK

(ground only)

1 40 9 1

N2 Maximum number of successive T1 timer expirations

1 20 3* 1

* For T1 (ground) = 2 and T1 (aircraft) = 10.

Table 5-12. Address type options

Code Description

000 Frame transmitted by the ground station to the specified aircraft (discretely addressed) - Radio Identifier = 1.

001 Frame transmitted by the specified aircraft to the ground station (discretely addressed) - Radio Identifier = 1.





010-110 Spare

111 Frame transmitted by the ground station to all aircraft, the 24-bit aircraft station address shall be set to all binary 1’s (broadcast).

Table 5-15a). XID parameters

Air-initiatedlink establishment

Link connection rejection

Source Aircraft New ground station

Any station

Destination Proposed ground station

Aircraft Any Peer station

XID parameters GI PI CTRL_CMD_LE CTRL_RSP_LE CTRL_RSP_LCR CTRL_CMD_LCR

Public parameters(8885:1993:Mode 3)

Parameter set ID 80h 01h N/A O N/A

N1-downlink 80h 05h N/A O N/A

N1-uplink 80h 06h N/A O N/A

Counter N2 80h 0Ah N/A O N/A

Timer T1 80h 09h N/A O N/A

T-ack 80h 08h N/A O N/A




Any station




Private parameters (V)

Parameter set ID F0h 00h O O M

SQP F0h 02h O O N/A

LCR cause F0h 06h N/A N/A M

Modulation support F0h 81h O N/A N/A

Private parameters (W)

Parameter set ID F0h 00h M M O

Algorithm version number F0h 01h O O O

Network initialization F0h 02h M M N/A

Connection check F0h 03h N/A N/A N/A

Version F0h 04h O O O

Authentication F0h 05h O O O

Subnetwork user data F0h 06h O O O

Assigned Altitude F0h 07h O O O

NM1 F0h 42h N/A O N/A

RR F0h 43h N/A O N/A

WR F0h 44h N/A O N/A

RE F0h 46h N/A O N/A

RL F0h 47h N/A O N/A

Timer T3 F0h 48h N/A O N/A

NL1 F0h 49h N/A O N/A

Timer TL3 F0h 4Ah N/A O N/A

Timer TL4 F0h 4Bh N/A O N/A

WE F0h 4Ch N/A O N/A

Operative_GNIp F0h 4Fh O O N/A

Aircraft_TMbB F0h 50h O O N/A

MbB_Operations_Permitted F0h 51h O O N/A

Expedited recovery F0h 80h O N/A N/A




Any station




t F0h 81h O N/A N/A

f F0h 82h O N/A N/A

Former_GNIp F0h 83h O O N/A

Counter NL2 F0h C1h N/A O N/A

NL5 F0h C2h N/A O N/A

NL6 F0h C3h N/A O N/A

Key for Table 5-15a) above:

Abbreviations used:

GI = ISO 8885 Group IdentifierPI = ISO 8885 Parameter IdentifierM = MandatoryO = OptionalN/A = Not applicableh = hexadecimal

Table 5-15b). XID parameters

Addressed link parameter modification Broadcast link parameter

modification

Source Current groundAny station

AircraftPeer station Current ground station

Destination AircraftPeer station Current groundOriginating

station

Aircraft

XID parameters GI PI CTRL_CMD_LPM CTRL_RSP_LPM CTRL_CMD_LPM(GSIF)

Public parameters (8885:1993:Mode3)

Parameter set ID 80h 01h O N/AO O

N1-downlink 80h 05h O N/AO O


modification




station

Aircraft


N1-uplink 80h 06h O N/AO O

Counter N2 80h 0Ah O N/AO O

Timer T1 80h 09h O N/AO O

T-ack 80h 08h N/AO 0O N/A

Private parameters (V)

Parameter set ID F0h 00h O O O

SQP F0h 02h O O O

LCR cause F0h 06h N/A N/A N/A

Modulation support F0h 81h N/OA N/A N/A

Private parameters (W)

Parameter set ID F0h 00h O O O

Algorithm version number F0h 01h O N/AO O

Network initialization F0h 02h O O O

Connection check F0h 03h O O O

Version F0h 04h N/A N/A O

Authentication F0h 05h O O O

Subnetwork user data F0h 06h O O O

Assigned Altitude F0h 07h O O O

NM1 F0h 42h O N/AO O

RR F0h 43h O N/AO O

WR F0h 44h O N/AO O

RE F0h 46h O N/AO O

RL F0h 47h O N/AO O

Timer T3 F0h 48h O N/AO O

NL1 F0h 49h O N/A O

Timer TL3 F0h 4Ah O N/AO O


modification




station

Aircraft


Timer TL4 F0h 4Bh O N/AO O

WE F0h 4Ch O N/AO O

Operative_GNIp F0h 4Fh O O O

Aircraft_TMbB F0h 50h O O O

MbB_Operations_Permitted

F0h 51h O O O

Expedited recovery F0h 80h N/A N/A N/A

t F0h 81h OA N/A N/AO

f F0h 82h OA N/A N/AO

Former_GNIp F0h 83h OA O N/AO

Counter NL2 F0h C1h OG N/A O

NL5 F0h C2h OG N/A O

NL6 F0h C3h OG N/A O

Key for Table 5-15b) above:

Abbreviations used:

GI = ISO 8885 Group IdentifierPI = ISO 8885 Parameter IdentifierM = MandatoryO = Optional

OA = Optional for Aircraft station OG = Optional for Ground station

N/A = Not applicableh = hexadecimal

Table 5-27b). Network type values

Network Type Protocol

0000 0000 ATN/ISO_8208/VDL Mode 3 PLP compression

0000 0001 ATN/ISO_8473/ISO 9542/VDL Mode 3 CLNP compression

0000 0010 ATN/ISO_8208 (No subnetwork compression)Reserved

0000 0011 ATN/ISO_8473/ISO 9542 (No subnetwork compression)Reserved

0000 0100 Raw/VDL

0000 0101 ATN frame mode

0000 0101 to 1111 1111 Reserved

Table 5-29. Expedited recovery parameter

Parameter ID 1 0 0 0 0 0 0 0 Expedited recovery

Parameter length 0 0 0 0 0 1 0 1

Parameter value v8 v7 v6 v5 v4 v3 v2 v1 Local identifier

a24 a23 a22 a21 a20 a19 a18 a17 Aircraft address

a16 a15 a14 a13 a12 a11 a10 a9

a8 a7 a6 a5 a4 a3 a2 a1

r20 r10 p2 p1 s4 s3 s2 s1 Reservation Status

Table 5-38. NL1 parameterRESERVED

Parameter ID 0 1 0 0 1 0 0 1 NL1


Parameter value v8 v7 v6 v5 v4 v3 v2 v1 value

Table 5-46. NL5 parameter



Parameter value v8 v7 v6 v5 v4 v3 v2 v1 value

Table 5-47. NL6 parameter



Parameter value s 0 v6 v5 v4 v3 v2 v1 value

Table 6 - 2 : IPI and EIPI Protocol Identifiers

Protocol IPI EIPI

ACARS over VDL3 0xFF 0xFF

FIS-B 0xFF 0xFE

Reserved 0xFF 0x00

8.2.1 Speech encoding algorithm

The Voice Unit shall incorporate and default to the Advanced Multiband Excitation (AMBE) vocoder algorithm, version AMBE-ATC-10, from Digital Voice Systems, Incorporated (DVSI) for speech compression unless commanded otherwise.

Note 1.— The AMBE algorithm is subject to patents. See SARPs, Annex 10, Volume III, Part 1, Chapter 6 for more details.

Note 2.— The FEC parity bits are included in the 4800 bps transmission rate.

Note 3.— The algorithm must use a voice frame size compatible with the TDMA frame of 120 ms.

8.2.2 Voice burst framing

Exactly six voice frames shall be transmitted in a single V/D (voice) burst during full rate operations (TS1 and TS3). Exactly five voice frames shall be transmitted in a single V/D (voice) burst during truncated voice mode operations (TS2).

8.2.3 Vocoder frame bit ordering

8.2.3.1The 96 bits of the vocoder frame shall be ordered as follows:

- 114 –

For the DVSI VC-20 ATC-10B implementation, the vocoder encoder frame shall consist of the concatenation of 4-bit words output from the vocoder board, each transmitted msb first. The VC-20 ATC-10B vocoder decoder frame shall consist of the concatenation of twenty-four 4-bit words input to the vocoder board, each received msb first. For the VC-20 ATC-10B implementation, the 4-bit control words which precede each vocoder frame shall not be considered part of the 96-bit vocoder frame.

8.2.3.2All other implementations shall conform to this bit ordering.

Note.— Implementers are advised to contact Digital Voice Systems, Incorporated to ensure correct bit ordering if additional guidance is required.

8.2.3.3 The output of the vocoder implementation shall demonstrate bit exact compliance with output of the DVSI fixed-point C reference implementation when using the DVSI test vectors.

Note.— The Digital Voice Systems, Incorporated part number, for the test vectors and the output results, is 03-168-1229 rev 1.0 or part number as approved by the regulatory authority . The proper bit ordering is nibble-wise inverted from what DVSI will indicate as the normal bit ordering.


Aircraft Receiver State Squelch Window (Note 1) Ground ReceiverState SquelchWindow (Note 1)

StateMessage Type

TS1(Note 2)

TS2(Note 2, Note 3)

TS3 (Note 5)

TS1N/A(Note 2)




Downlink voice: TS3 Wide Open(Note 4)

Wide Open(Note 4)

Wide Open Wide Open(Note 4)

Note 1.— Each squelch window is defined relative to the nominal burst time in symbols.Note 2.— n = squelch window parameter.Note 3.— If n is unknown, then the default value is 6. N=7 is not a valid Squelch Window parameter.Note 4.— Receive any such message with “correct” group ID if not already receiving a voice message.Note 5.— In this column, “Wide Open” implies that an aircraft radio will receive voice messages without regard to Group ID based on the rules defined in Section 5.2.4.2. 2 5 .


- 115 –


Aircraft Receiver State Squelch Window (Note 1) Ground Receiver State Squelch Window (Note 1)

StateMessage Type

TS1(Note 2)

TS2(Note 2, Note 3)

TS3 (Note 5)

TS1N/A(Note 2)




Downlink voice: TS3 Wide Open(Note 4)

Wide Open(Note 4)

Wide Open Wide Open(Note 4)

Note 1.— Each squelch window is defined relative to the nominal burst time in symbols.Note 2.— n = squelch window parameter.Note 3.— If n is unknown, then the default value is 7.Note 4.— Receive any such message with “correct” group ID if not already receiving a voice message.Note 5.— In this column, “Wide Open” implies that an aircraft radio will receive voice messages without regard to Group ID based on the rules defined in Section 5.2.4.2. 5 2 .

FIGURES FOR THE MANUAL ON VHF DIGITAL LINK (VDL) MODE 3 TECHNICAL SPECIFICATIONS


APPENDIX B — FORMAT AND USAGE OF THE SYSTEMDATA AND HEADER SEGMENTS FOR MODE 3 OPERATION

2.1.1The non-3T configuration uplink System Data segment consists of 48 bits or 6 octets. The first three octets, including Message ID, System Configuration, Voice Signal, Aircraft ID (poll), BeaconSlot ID, Ground Station Code, and Squelch Window, are common for all uplink messages. The remaining three octets are assigned depending on the message type of the M uplink burst. Figure B-1 displays the non-3T configuration generic uplink System Data segment format.

First bit transmitted

Bit Number

Octet No. 1 2 3 4 5 6 7 8

1 Message ID Voice signal Aircraft ID (Poll)

2 Aircraft ID (Poll) (cont.) System Configuration

3 Slot Beacon ID Ground Station Code Squelch Window

4

(Message Dependent)5

6

Figure B-1. Non-3T configuration generic M uplink system data segment

2.1.2Message ID. The coding of the Message ID field is given in Table B-1. This field is used to aid the receiver in identifying the burst type.

Table B-1. M burst messagesMessage ID M bursts Paragraph No.

of this appendix

Uplink 0000 Normal message 2.2

0001 Net entry response message (no previous link)

2.3

0010 Net entry response message (previous link preserved)

2.3

0011 Next net command message 2.4

0100 Recovery message 2.5

0101 Handoff check message 2.6

0110 Terminate Net message 2.14

0111 Supported Options message 2.15

Message ID M bursts Paragraph No. of this

appendix

Downlink Identified by separate sync

sequence

Net entry request message 2.9

1000* Reservation request message 2.10

1001* Acknowledgement message 2.11

1010* Leaving net message 2.12

1011-1111* Spare

1000** 4-slot configurationV/3V1D poll response

2.10

1001** 2V2D poll responseSpare 2.10

1010** 3-slot configuration1V3D poll response

2.10

1011** 3T poll responseSpare 2.10

1100** 3V/3S/2S1XT configuration poll response

2.10

1101** Spare 2V1D poll response 2.10

1110** Spare 1V2D poll response 2.10

1111** Next net ACK 2.13

* Identified by standard synchronization sequence defined in Annex 10 Volume III, Part I, Chapter 6.** Identified by special synchronization sequence for poll responses defined in Annex 10 Volume III, Part I, Chapter

6.

Note 1.— For system configurations with dedicated Poll Response message ID, e.g., 2V2D and 1V3D, the Poll Response message uniquely identifies the system configuration.

Note 2.— For systems configurations that share the same Poll Response message ID, e.g., 4V and 3V1D, the Poll Response message alone may not uniquely identify the system configuration. Nevertheless, the aircraft radio can infer the position of the common slots that are allowed for voice transmissions, based on the allowable system configurations and the slot in which the Poll Response is being transmitted.

2.1.5Aircraft ID (poll). The Aircraft ID is a 6-bit field used to identify uniquely aircraft of a given user group. It is assigned by the ground station as part of the net entry process. The valid Aircraft IDs are listed in Table B-4:

Table B-4. Aircraft ID codes

Code Description

0 Ground

1-60 Aircraft user

61 Discrete addressing not supported

62 Discrete addressing not supported (use Alternative Validity Window)Spare

63 Broadcast

As part of the generic portion of the M uplink system data segment, this field is used to poll a selected aircraft. When polled, an aircraft transmits a M downlink message in the next MAC cycle without contention. The value of zero is used in Net Entry Response messages to indicate that the net is full.

2.1.6BeaconSlot ID. For all system configurations except 2V2D, 2V1D, 1V3D and 1V2D, tThe slot beacon ID indicates the slot the message is associated with in order to provide the VDL receiver with proper TDMA frame alignment. The valid codes are described in Table B-5.

2.1.6.1 For system configurations 2V2D, 2V1D, 1V3D and 1V2D, the Beacon ID field indicates the Local ID prefix of the aircraft radio being polled by that beacon signal.

Note.— For these configurations the beacon always occurs in slot A, so there is no time ambiguity to resolve when the configuration is known. (The system configuration information is part of the same Golay code word as the Beacon ID.).

Table B-5. Slot Beacon ID codes

Code Description

00 Slot A

01 Slot B

10 Slot C

11 Slot D

2.1.8Squelch window. The Squelch Window provides a mechanism to suppress certain received voice messages as co-channel interference based on the time of arrival of the V/D (voice) bursts. The Squelch Window specifies the range, expressed in symbol periods, within which the first symbol of the synchronization sequence of a voice burst must occur. Voice bursts with the first symbol of the synchronization sequence arriving outside the Squelch Window shall be suppressed. The Squelch Window parameter n is equal to the numerical value of the 3-bit Squelch Window field. To find the Squelch Window range for a given Squelch Window code, substitute the value n in Table 8-1 and

Table 8-2 for the corresponding system configurations, aircraft VDL radio timing states, and the message type of the incoming voice traffic.

2.2.2Each Reservation Response field can signal the actual allocation of channel resources to a requesting aircraft.


Bit Number

Octet No. 1 2 3 4 5 6 7 8

1 Message ID Voice Signal Aircraft ID (Poll)



4 Reservation Response #1

5 Reservation Response #1 (cont.) Reservation Response #2

6 Reservation Response #2 (cont.)

Figure B-2. M uplink (normal message)

2.2.3Reservation response. Each Reservation Response field contains three subfields as follows:

Group ID (2 bits): The Group ID is coded identically as the Slot Beacon ID (2.1.6 of this Appendix). In this case it is used in conjunction with the Aircraft ID (listed below) to identify uniquely the aircraft across all user groups supported on the frequency by the ground station.

Note.— The Group ID and the Aircraft ID fields are collectively referred to as the Local User ID.

Aircraft ID (6 bits): 2.1.5 of this Appendix defines the valid codes. When the Reservation Response field is not used, the value of 0 shall be used in the Aircraft ID field.

Reserved Slot (4 bits): This subfield indicates the slot for which V/D (data) burst access has been scheduled to begin. This subfield is coded as indicated in Table B-6:

Table B-6. Reserved slot subfield encoding

Code Description

0000 No Reservation Information

0001 RACK (UDR Supported)

0010 RACK (UDR Not Supported)

0011 Spare

0100 Data Reservation (start in even B-slot)

0101 Data Reservation (start in odd B-slot)

0110 Data Reservation (start in even B-slot, skip D)

Code Description

0111 Data Reservation (start in odd B-slot, skip D)

1000 Data Reservation (start in even C-slot)

1001 Data Reservation (start in odd C-slot)

1010 Data Reservation (start in even C-slot, skip D)

1011 Data Reservation (start in odd C-slot, skip D)

1100 Data Reservation (start in even D-slot)

1101 Data Reservation (start in odd D-slot)

1110 Voice Reservation (start in even D-slot)

1111 Voice Reservation (start in odd D-slot)

Note.— For non-3T configurations, use only the codes with the third bit set to 0. For system configuration 2V2D, allowable values are 0000, 0001, 1000, 1001, 1100 and 1101. For system configuration 3V1D, allowable values are 0000, 0001, 1100 and 1101. For system configuration 2V1D, allowable values are 0000, 0001, 1000 and 1001. For configuration 1V3D the allowable values are 0000, 0001, 0100, 0101, 1000, 1001, 1100, and 1101. For configuration 1V2D the allowable values are 0000, 0001, 0100, 0101, 1000, and 1001.

2.3 Net entry response message

This message type is used to acknowledge the Net Entry Request message and to assign the Local User ID to the requesting aircraft. This Local User ID is the combination of the Slot Beacon ID field and the Aircraft ID field. The aircraft ICAO address is attached to allow the receiver to validate the message. This message is shown in Figure B-3.


Bit Number

Octet No. 1 2 3 4 5 6 7 8




4ICAO Aircraft Address5

6

Figure B-3. M Uplink (Net Entry Response Message)

Note.— Two separate Message IDs are used for this message. If the new ground station is unable to redirect the link from the previous ground station Message ID = 0001, “no previous link,” is used. If the new ground station can support redirection of the link from the previous ground station Message ID = 0010, “previous link preserved” is used.

2.4.4Net type. This field indicates whether the next net is a TDMA net or simply an AM frequency. A 0 indicates AM and a 1 indicates TDMA.

First bit Bit Numbertransmitted

Octet No. 1 2 3 4 5 6 7 8




4 Next Net

5 Next Net (continue) Net Type Next Ground Station Code

6 Local User ID

Figure B-4. M Uplink (Next Net Command Message)

2.5 Recovery message

This message is used to inform aircraft that the ground station is in the link recovery state. This message invokes recovery procedures in the aircraft LME as standardized in 5.4 of the Technical Specifications. This message is shown in Figure B-5.


Octet No. 1 2 3 4 5 6 7 8




4 Reservation Response #1

5 Reservation Response #1 (cont.) Reservation Response #2

6 Reservation Response #2 (cont.)

Figure B-5. M Uplink (Recovery Message)

2.7 3T Configuration M uplink system data segment (generic elements)

The 3T configuration uplink System Data segment consists of 192 bits or 24 octets. The fourth through the sixth semi-octets, including the System Configuration, BeaconSlot ID, Ground Station Code, and Squelch Window fields comprising the fourth through the sixth semi-octets, are common to all uplink messages. (Because polling is controlled by other fields in the uplink message, the Beacon ID field is always set to the value “A.”) Semi-octets 1 through 3 and 7 through 12 are used to control activities in User Group B. Semi-octets 1 through 3 control polling access for users in User Group B. Semi-octets 7 through 12 can be Net Entry Response messages, Next Net Command message, or Recovery message. Similarly, semi-octets 13 through 21 are devoted to User Group C; semi-octets 22 through 30 are devoted to User Group D. Figure B-7 displays the 3T configuration generic uplink System Data segment format.


Octet No. Semi-octet 1 2 3 4 5 6 7 8

1 1, 2 Message ID (B) Voice Signal (B) Aircraft ID (Poll) (B)

2 3, 4 Aircraft ID (Poll) (B) (cont.) System Configuration

3 5, 6 Slot Beacon ID Ground Station Code Squelch Window

4 7, 8

Group B Message5 9, 10

6 11, 12

7 13, 14 Message ID (C) Voice Signal (C) Aircraft ID (Poll) (C)

8 15, 16 Adjacent frequency 4 (cont.)

9 17, 18Group C Message

10 19, 20

11 21, 22 Message ID (D)

12 23, 24 Voice Signal (D) Aircraft ID (Poll) (D)

13 25, 26

Group D Message14 27, 28

15 29, 30

16 31, 32 Reservation #1

17 33, 34 Reservation #1 (cont.) Reservation #2

18 35, 36 Reservation #2 (cont.)







Figure B-7. 3T Configuration Generic M Uplink System Data Segment

2.12 Leaving net message

The Leaving Net message is shown in Figure B-11. It is used by the aircraft to notify the ground station that the aircraft is leaving the net. Note that in order to minimize the probability of false link disconnection, the 9 trailing zeros in the Leaving Net message are used by the receiving ground station for verification (see Section 5.4.2.6.1).

First bit Bit NumberTransmitted

Octet No. 1 2 3 4 5 6 7 8

1 Message ID Local User ID

2 Local User ID (cont.) Ground Station Code Spare0

3 0Spare (cont.)

0 0 0 0 0 0Spare 0

Figure B-11. M Downlink (Leaving Net Message)

2.15 Supported Options Message

The Supported Options message shown in Figure B-12b, is used to indicate options that are supported by the current ground station. One Reservation Response field is included in this message, along with an Options field. Each Reservation Response field consists of 12 bits. The Reservation Response field can signal the actual allocation of channel resources to the entering aircraft. The Options field is a 12-bit field with each bit set to 1 to indicate a specific option is supported and 0 if not supported, as described in Table B-9. Reserved bits are set to 0. The Radio Identifier field is described in Table B-10.

first bit transmitted

Bit Number

Octet No. 1 2 3 4 5 6 7 8



3 Beacon ID Ground Station Code Squelch Window

4 Reservation Response

5 Reservation Response (cont.)Options

6Options (cont.)

Figure B-12b: M Uplink (Supported Options Message)

Table B-9: Options field

Octet Bit Number Field5 5 Urgent Downlink Request5 6-8 Reserved6 13-8 Reserved6 1-2 Radio Identifier

Table B-10: Radio Identifier field

Bit 2 Bit 1 Field0 0 Radio Identifier #10 1 Radio Identifier #21 0 Radio Identifier #31 1 Unavailable

3.1.1Message ID. This field is used to identify V/D bursts. The coding of this field is shown in Table B-9.

Table B-9. V/D burst type coding

Message ID V/D Burst Type

000 Uplink voice

001 Downlink voice: TS1



100 Uplink data – Acknowledged

101 Downlink data – Acknowledged

0110 Uplink data – Unacknowledged

111 Downlink data – Unacknowledged

— — — — — — — —

APPENDIX C — RANDOM NUMBER GENERATION ALGORITHM

1. Certain functions of the radio system related to RA require the generation of a “random” number. For instance, in all configurations the radio normally must choose which slot to use for a Net Entry Request. To ensure that different radios generate different numbers, the algorithm for generating random numbers is based on the unique 24-bit ICAO address. Because it is possible that on some platforms there may be more than one radio using the same ICAO address, the algorithm also makes use of a radio ID to allow each radio on an aircraft to have an independent random number generator. One example of such a radio ID is the 2-bit Source/Destination Identifier (SDI) currently used in a multiple radio installation on some platforms to identify the left radio, the center radio, the right radio, and all radios (all call). The 2-bit radio ID is set to 00 whenever there is only one radio on board a particular aircraft.

The algorithm generates a new 12-bit number at each access as follows. First, the 24-bit address is broken into 2 12-bit pieces ( and ) so that

These are used to generate two sequences of numbers using the recursion relations

N(n) = ((4021 + SDI) N(n-1) + a(1) +1) mod 4099

with M(0) =0 and N(0) = SDI. These are then combined to give a single random number

.

This ensures that two radios will never choose the same pseudorandom number repeatedly.

This algorithm produces 224 unique sequences, each of which is nearly uniform in distribution. No two sequences have any long subsequences in common, no matter what their relative time offset might be. These sequences can be used to derive sequences of numbers drawn from a smaller alphabet. Suppose, for example, that a sequence of numbers between 0 and 8 (inclusive) must be generated for random access in configuration 2V2D. Such a number can be derived from by invoking the formula

where means the integer part of the argument.

— — — — — — — —

APPENDIX D — RANDOM ACCESS DOWNLINK M BURSTSELECTION ALGORITHMS FOR AIRCRAFT RADIO

The method for selecting an appropriate random access downlink M burst to transmit a Net Entry Request (E), a Reservation Request (R), and a Leaving Net (L) type message shall be either one of the two algorithms described below. The process for selecting an appropriate downlink M burst for subsequent retransmission of a Net Entry message and a Reservation Request message shall also be as described below.

PREFERRED METHOD

The preferred method allows for faster access and results in shorter throughput delays in most network loading scenarios.

Step 1. Request for a RA downlink M burst arrives at MAC layer prior to LBACx in MAC cycle N.

Initial Transmission

Step 2. Transmit in the next available RA downlink M burst, e.g., in LBACy in MAC cycle Z.

Step 3. For Net Entry and Reservation Request, the aircraft radio shall wait for Wn (n = E or R) MAC cycles, respectively, for an appropriate reply from the ground station until LBACy in MAC cycle Z+Wn before attempting to retransmit the net entry or reservation request message. The parameter Wn shall be a MAC Service (or LME) System Parameter dependent on configuration and process type (n = E or R), i.e., WE for Net Entry and WR for Reservation Request. If a valid response is received on or before LBACy in MAC cycle Z+Wn the process shall be terminated.

Retransmission

Step 4. If a valid response is not received by LBACy in MAC cycle Z + Wn, where Z and Wn are defined in Step 3, the Net Entry or Reservation Request message shall be retransmitted using the procedures beginning in Step 5.

Step 5. Aircraft radio shall choose a random integer, r, which obeys 0 ≤ r ≤ Rn, using the algorithm of Appendix C. Rn shall be a MAC Service (or LME) System Parameter dependent on configuration and process type (n = E or R), i.e., RE for Net Entry and RR for Reservation Request.

Step 6. Aircraft radio shall skip r available RA opportunities and transmit in the next available RA downlink M burst and then return to Step 3.

Note 1.— The available LBACs depend on process type (Net Entry, Reservation Request, or Leave Net) and configuration. LBACs scheduled for Poll Responses and downlink ACKs are removed from the lists. The initial list includes all downlink M burst LBACs available to the relevant net.

Note 2.— Steps 3 and beyond are not applicable to Leaving Net Messages since a Leaving Net message requires no reply.

Note 3.— Failure to detect an uplink V/D (data) burst with an EOM indication may cause an aircraft radio to transmit a random access burst in conflict with a downlink ACK transmission from another aircraft radio. Duplicate messages may occur as a result.

ALTERNATE METHOD

Although the previous algorithm is preferred, the following method is also allowed. The alternate method is similar to the preferred method except that an initial randomization process is included.

Step 1. Request for a RA downlink M burst arrives at MAC layer prior to LBACx in MAC cycle N.

Initial Transmission

Step 2. Aircraft radio shall choose a random integer, r, which obeys 0 ≤ r ≤ Rn, using algorithm of Appendix C. Rn shall be a MAC Service System (or LME) Parameter dependent on configuration and process type (n = E, R, or L), i.e. RE for Net Entry, RR for Reservation Request, and RL for Leaving Net.

Step 3. Aircraft radio shall skip r available RA opportunities and transmit in the next available RA downlink M burst, e.g. in LBACy in MAC cycle Z.

Step 4. For Net Entry and Reservation Request, the aircraft radio shall wait for Wn (n = E or R) MAC cycles, respectively, for appropriate reply from the ground station until LBACy in MAC cycle Z+Wn before attempting to retransmit the net entry or reservation request message. The parameter Wn shall be a MAC Service (or LME) System Parameter dependent on configuration and process type (n = E or R), i.e. WE for Net Entry and WR for Reservation Request. If a valid response is received on or before LBACy in MAC cycle Z+Wn the process shall be terminated.

Retransmission

Step 5. If a valid response is not received by LBACy in MAC cycle Z + Wn, where Z and Wn are defined in Step 4, the Net Entry or Reservation Request message shall be retransmitted by repeating the above procedures beginning in Step 2.

Note 1.— The available LBACs depend on process type (Net Entry, Reservation Request or Leave Net) and configuration. LBACs scheduled for Poll Responses and downlink ACKs are removed from the lists. The initial list includes all downlink M burst LBACs available to the relevant net.

Note 2.— Steps 4 and 5 are not applicable to Leaving Net since a Leaving Net message requires no reply.

Note 3.— Failure to detect a downlink V/D (data) burst with an EOM indication may cause an aircraft VDL to transmit in an uplink RA M channel, which has been reserved for downlink ACK for another aircraft and corrupt the downlink ACK transmission. Duplicate messages will occur as a result.

— — — — — — — —

APPENDIX E — ISO 8208 COMPRESSION DEFINITION

1.2 ISO 8208 compression header definition

ISO 8208 Compression header definition shall be defined as protocol type field 0000 (binary) indicating ISO 8208 protocol and Compression type field (Type-specific field) 0001 (binary) indicating VPLP compression.

2.3 Call initiation

High priority virtual circuit calls shall be air-initiated. Low priority virtual circuit calls may be either ground or air-initiated.

3.3 Priority

3.3.1 Low priority calls. The priority facility shall encode low priority calls using the value of 0.

3.3.2 High priority calls. The priority facility shall encode high priority calls using the value of 2.

4.1 State transitions

The DCE shall operate as a state machine with the states as identified in Figure E.4-1. Upon entering a state, the DCE shall perform the actions specified in Table E.6-1. If a state transition is specified, the actions taken shall be as specified in Tables E.11-1 through E.11-109.

4.2 Disposition of packets

4.2.1Receipt from DTE. Upon receipt of a packet from the DTE, the packet shall be forwarded or not forwarded to the XDCE (GDCE/ADCE) (via the internetworking function) according to the parenthetical instructions contained in Tables E.11-1 to E.11-6. If no parenthetical instruction is listed or if the parenthetical instruction indicates “do not forward”, the packet shall be discarded.

4.2.2Receipt from XDCE. Upon receipt of a packet from the XDCE (via the internetworking function), the packet shall be forwarded or not forwarded to the DTE according to the parenthetical instructions contained in Tables E.11-7 to E.11-910. If no parenthetical instruction is listed or if the parenthetical instruction indicates “do not forward”, the packet shall be discarded.

READY AND RESTART STATES r1

r2

r3

CALL SETUP AND CLEARING STATES p1

p2

p3

p4

p5

p6

p76

DATA TRANSFER STATES d1

d2

d3

INTERRUPT & CONTROL STATES f1

f2

g1

g2

i1

i2

j1

j2

Note.— State r1, p4 and d1 are states that provide to the lower level of the DCE state hierarchy.

Figure E.4-1. DCE sub-state hierarchy

5.2.8 INTERRUPT RESERVED

5.2.8.1 Translation into VDL Mode 3 packets

5.2.8.1.1General. Reception by the XNI Internetworking Process of an ISO 8208 INTERRUPT Packet from the local DCE shall result in the generation of a corresponding VDL Mode 3 INTERRUPT Packet as specified in Table E.5-8:

Table E.5-8: VDL Mode 3 INTERRUPT packet

MSB Bit Number LSB

Octet 8 7 6 5 4 3 2 1

1 DP VP SP ST FILL

2 FILL SN

3 CH FILL

4 + n UDv

Note.— Fields shown in the packet format and not specified in the following paragraphs shall be set as specified in 5.2.1 of this Appendix.

5.2.8.1.2Data packet type (DP). This field shall be set to 0.

5.2.8.1.3VSP Packet Type (VP). This field shall be set to 1.

5.2.8.1.4Supervisory packet (SP). This field shall be set to 3.

5.2.8.1.5Supervisory type (ST). This field shall be set to 1.

5.2.8.1.6User data (UD). The user data shall be transferred from the ISO 8208 packet to the VDL Mode 3 packet.

5.2.8.2Translation into ISO 8208 packets. Reception by the XNI Internetworking process of a VDL Mode 3 INTERRUPT Packet from the local XDCE shall result in the generation of a corresponding ISO 8208 INTERRUPT Packet to the local DCE. The translation from the VDL Mode 3 packet to the ISO 8208 packet shall be the reverse of the processing defined in 5.2.8.1 of this Appendix.

5.2.9 INTERRUPT CONFIRMATIONRESERVED

5.2.9.1 Translation into VDL Mode 3 packets

5.2.9.1.1General. Reception by the XNI Internetworking Process of an ISO 8208 INTERRUPT CONFIRMATION Packet from the local DCE shall result in the generation of a corresponding VDL Mode 3 INTERRUPT CONFIRMATION Packet as specified in Table E.5-9:

Table E.5-9: VDL Mode 3 INTERRUPT packet

MSB Bit Number LSB

Octet 8 7 6 5 4 3 2 1

1 DP VP SP ST SS

2 FILL SN

3 CH FILL

Note.— Fields shown in the packet format and not specified in the following paragraphs shall be set as specified in 5.2.1 of this Appendix.

5.2.9.1.2Data packet type (DP). This field shall be set to 0.

5.2.9.1.3VSP packet type (VP). This field shall be set to 1.

5.2.9.1.4Supervisory packet (SP). This field shall be set to 3.

5.2.9.1.5Supervisory type (ST). This field shall be set to 3.

5.2.9.1.6Supervisory subset (SS). This field shall be set to 2.

5.2.9.2Translation into ISO 8208 packets. Reception by the XNI Internetworking process of a VDL Mode 3 INTERRUPT CONFIRMATION Packet from the local XDCE shall result in the generation of a corresponding ISO 8208 INTERRUPT CONFIRMATION Packet to the local DCE. The translation from the VDL Mode 3 packet to the ISO 8208 packet shall be the reverse of the processing defined in 5.2.9.1 of this Appendix.

6.1 State transitions

The XDCE shall operate as a state machine. Upon entering a state, the XDCE shall perform the actions specified in Table E.6-1. If a state transition is specified, the actions taken shall be as specified in Tables E.11-11 through E.11-1817.

Note 1.— The next state transition (if any) that occurs when the XDCE receives a packet from the peer XDCE is specified by Table E.11-11 through Table E.11-15. The same transitions are defined in Table E.11-16 through Table E.11–187 when the XDCE receives a packet from the DCE (via the Internetworking Process).

Note 2.— The XDCE state hierarchy is the same as for the DCE as presented in Figure E.4-1, except that states r2, r3 and p5 are omitted.

6.3.5Clear collision. A clear collision occurs at the XDCE when it receives a CLEAR REQUEST Packet from the DCE and then receives a CLEAR REQUEST Packet from the peer XDCE (or vice versa). In this event the XDCE does not expect to receive a CLEAR CONFIRMATION Packet for this SVC and shall consider the clearing complete.

Table E.6-1. DCE actions at state transition

DCESTATE

STATE DEFINITION ACTION THAT SHALL BE TAKEN WHEN ENTERING THE STATE

r1 PACKET LEVEL READY

Return all SVCs to the p1 state (see p1 state explanation).

r2 DTE RESTART REQUEST

Return each SVC to the p1 state (see p1 state explanation). Issue a RESTART CONFIRMATION to the DTE.

r3 DCE RESTART REQUEST

Issue a RESTART REQUEST to the DTE. Unless entered via the r2 state, send a RESTART REQUEST to the internetworking process.

P1 READY Release all resources assigned to SVC. Break the correspondence between the DTE/DCE SVC and the ADCE/GDCE SVC (the ADCE/GDCE SVC may not yet be in the p1 state).

P2 DTE CALL REQUEST Determine if sufficient resources exist to support the request; if so, allocate resources and forward a CALL REQUEST packet to the internetworking process; if not, enter DCE CLEAR REQUEST to DTE state (p7). Determination of resources and allocation is as defined in ISO 8208.

P3 DCE CALL REQUEST Determine if sufficient resources exist to support the request; if so, allocate resources and forward a CALL REQUEST packet to the DTE; if not, send a CLEAR REQUEST packet to the reformatting process. Determination of resources and allocation is as defined in ISO 8208.

P4 DATA TRANSFER No action.

P5 CALL COLLISION Reassign outgoing call to another SVC (the DTE in its call collision state ignores the incoming call) and enters the DCE CALL REQUEST state (p3) for that new SVC.

P6 DTE CLEAR REQUEST Release all resources assigned to the SVC, send a CLEAR CONFIRMATION packet to the DTE and enter p1 state.

P7 DCE CLEAR REQUEST to DTE

Forward a CLEAR REQUEST packet to DTE.

d1 FLOW CONTROL READY

No action.

d2 DTE RESET REQUEST Remove Data Packets transmitted to the DTE from window; discard any DATA packets that represent partially transmitted M-bit sequences and discard any INTERRUPT packet awaiting transfer to the DTE; reset all window counters to zero; set any timers and retransmission parameters relating to DATA and INTERRUPT transfer to their initial value. Send RESET CONFIRMATION packet to DTE. Return SVC to d1 state.

d3 DCE RESET REQUEST to DTE

Remove Data Packets transmitted to the DTE from window; discard any DATA packets that represent partially transmitted M-bit sequences and discard any INTERRUPT packet awaiting transfer to the DTE; reset all window counters to zero; set any timers and retransmission parameters relating to DATA and INTERRUPT transfer to their initial value. Forward RESET REQUEST packet to DTE.

DCESTATE

STATE DEFINITION ACTION THAT SHALL BE TAKEN WHEN ENTERING THE STATE

i1 DTE INTERRUPT READY

No action.

i1 DTE INTERRUPT SEND Forward INTERRUPT packet received from DTE to internetworking process.

j1 DCE INTERRUPT READY

No action.

j2 DCE INTERRUPT SENT Forward INTERRUPT packet received from internetworking process to DTE.

f1 DCE RECEIVE READY No action.

f2 DCE RECEIVE NOT READY

No action.

g1 DTE RECEIVE READY No action.

g2 DTE RECEIVE NOT READY

No action.

6.4 Data transfer and interrupt procedures

6.4.1 General

6.4.1.1Independent SVC. Data transfer and interrupt procedures shall apply independently to each SVC. The contents of the User Data Field shall be passed transparently to the DCE or the peer XDCE. Data shall be transferred in the order dictated by the sequence numbers assigned to the data packets.

6.4.5 Interrupt procedures for switched virtual circuitsRESERVED

6.4.5.1Non-use of flow control. If user data is to be sent via the Mode 3 subnetwork without following the flow control procedures, the interrupt procedures shall be used. The interrupt procedure shall have no effect on the normal data packet and flow control procedures. The processing of the VDL Mode 3 INTERRUPT Packet shall occur as soon as it is received by the XDCE.

Note.— The use of clear, reset and restart procedures can cause interrupt data to be lost.

6.4.5.2Interrupt processing. Interrupt processing shall have precedence over any other processing from the SVC occurring at the time of the interrupt.

6.4.5.3Unconfirmed Interrupt. The reception of a VDL Mode 3 INTERRUPT Packet before the previous interrupt of the SVC has been confirmed (by the receipt of a VDL Mode 3 INTERRUPT CONFIRMATION Packet) shall be defined as an error.

6.7.1Reception of Reset Request. When the XDCE receives a VDL Mode 3 RESET REQUEST Packet from either the peer XDCE or the DCE or due to an error condition performs its own reset, the following actions shall be taken:

a) Those VDL Mode 3 DATA Packets that have been transmitted to the peer XDCE shall be removed from the window.

b) Those VDL Mode 3 DATA Packets that are not transmitted to the peer XDCE but are contained in an M-bit sequence for which some packets have been transmitted shall be deleted from the queue of DATA packets awaiting transmission.

c) Those VDL Mode 3 DATA Packets received from the peer XDCE that are part of an incomplete M-bit sequence shall be discarded.

d) The lower window edge shall be set to zero and the next packet sent shall have a sequence number (PS) of zero.

e) Any outstanding VDL Mode 3 INTERRUPT Packets to or from the peer XDCE shall be left unconfirmed.

f) Any Mode 3 INTERRUPT Packet awaiting transfer shall be discarded.

.

11.2 Diagnostic and cause codes

The Table entries for certain conditions indicate a diagnostic code that shall be included in the packet generated when entering the state indicated. The term, “D =,” shall define the diagnostic code. When “A = DIAG,” the action taken shall be to generate an ISO 8208 DIAGNOSTIC packet and transfer it to the DTE; the diagnostic code indicated shall define the entry in the diagnostic field of the packet. In the cause field of any packet type, bit 8 of the cause field shall always be set to zero, indicating that the condition was detected by the DTE/DCE interface. Unless otherwise specified, the clearing cause code and the reset cause code shall be set as specified in ISO 8208.

Note 1.— The tables provided below specify state requirements in the following order:

E.11-1DCE special cases

E.11-2DTE effect on DCE restart states

E.11-3DTE effect on DCE call setup and clearing states

E.11-4DTE effect on DCE reset states

E.11-5DTE effect on DCE interrupt transfer states

E.11-6DTE effect on DCE flow control transfer states

E.11-7XDCE effect on DCE restart states

E.11-8XDCE effect on DCE call setup and clearing states

E.11-9XDCE effect on DCE reset states

E.11-10XDCE effect on DCE interrupt transfer states

E.11-11GNI(ANI) effect on ADCE (GDCE) packet layer ready states

E.11-12GNI(ANI) effect on ADCE (GDCE) call setup and clearing states

E.11-13GNI(ANI) effect on ADCE (GDCE) reset states

E.11-14GNI(ANI) effect on ADCE (GDCE) interrupt transfer states

E.11-15GNI(ANI) effect on ADCE (GDCE) flow control transfer states

E.11-16DCE effect on ADCE (GDCE) call setup and clearing states

E.11-17DCE effect of ADCE (GDCE) reset states

E.11-18DCE effect on ADCE (GDCE) interrupt transfer states

Note 2.— All tables specify both ANI and GNI actions.

Note 3.— Within the VDL Mode 3 subnetwork, states p6 and d2 are transient states.

Note 4.— References to “notes” in the State Tables refer to table-specific notes that follow each State Table.

Note 5.— All diagnostic and cause codes are interrupted as decimal numbers.

Note 6.— An SVC between an ADCE and a GDCE may be identified by a temporary and/or permanent channel number, as defined in 5.1.1 of this Appendix.

Table E.11-3. DTE effect on DCE call setup and clearing states

DCE Call Setup and Clearing States (see Note 5)

Packet Receivedfrom DTE

READY

p1

DTE CALL REQUEST

p2

DCE CALL REQUEST

p3

DATA TRANSFER

p4

CALL COLLISION

p5

DTE CLEAR REQUEST

p6

DCE CLEAR REQUEST

to DTEp7

Packets having a packet type identifier shorter than 1 byte

A = ERRORS = p7D = 38

A = ERRORS = p7D = 38(see Note 2)


See Table J.11-4



A = DISCARD

Packets having a packet type identifier which is undefined or not supported by DCE




See Table J.11-4



A = DISCARD

RESTART REQUEST orRESTART CONFIRMATIONpacket with logical channel identifier unequal to 0




See Table J.11-4



A = DISCARD

CALL REQUEST A = NORMALS = p2(forward)


A = NORMALS = p5




A = DISCARD

CALL ACCEPT A = ERRORS = p7D = 41


A= NORMALS = p4orA = ERRORS = p7D = 42(see Notes 2 and 3)


A = ERRORS = p7D = 24(see Note 2 and 4)


A = DISCARD

CLEAR REQUEST A = NORMALS = p6

A = NORMALs = p6(forward)




A = DISCARD

A = NORMALS= p1(do not forward)

CLEAR CONFIRMATION







A = NORMALS = p1(do not forward)

DATA, flow control or reset packet




See Table J.11-4



A = DISCARD

DCE call setup and clearing states (see Note 5)

Packet received

from DTE

READY

p1

DTE CALL REQUEST

p2

DCE CALL REQUEST

p3

DATA TRANSFER

p4

CALL COLLISION

p5

DTE CLEAR REQUEST

p6

DCE CLEAR REQUEST to

DTEp7

Packets having a packet type identifier shorter than 1 byte




See Table E.11-4



A = DISCARD

Packets having a packet type identifier which is undefined or not supported by DCE




See Table E.11-4



A = DISCARD

RESTART REQUEST,RESTART


Packet received

from DTE

READY

p1

DTE CALL REQUEST

p2

DCE CALL REQUEST

p3

DATA TRANSFER

p4

CALL COLLISION

p5

DTE CLEAR REQUEST

p6


DTEp7

CONFIRMATIONorREGISTRATION packet with logical

channel identifier unequal to 0




See Table E.11-4

A = DISCARDA = ERRORA = ERROR

S = p7D = 41(see Note 2)

S = p7D = 41(see Note 2)

CALL REQUEST

A = NORMALS = p2(forward)


A = NORMALS = p5




A = DISCARD

CALL ACCEPT



A= NORMALS = p4orA = ERRORS = p7D = 42(see Notes 2 and 3)


A = ERRORS = p7D = 24(see Notes 2 and 4)


A = DISCARD

CLEAR REQUEST

A = NORMALS = p6





A = DISCARD


CLEAR CONFIRMATION






A = ERRORS = p7D = 25(see



Packet received

from DTE

READY

p1

DTE CALL REQUEST

p2

DCE CALL REQUEST

p3

DATA TRANSFER

p4

CALL COLLISION

p5

DTE CLEAR REQUEST

p6


DTEp7

Note 2)

DATA, interrupt, flow control or reset packet




See TableE.11-4



A = DISCARD

Notes.—

1. On entering the p5 state, the DCE reassigns the outgoing call to the DTE to another channel (no CLEAR REQUEST is issued) and responds to incoming the DTE call as appropriate with a CLEAR REQUEST of CALL ACCEPT packet.

2. The error procedure consists of performing the actions specified when entering the p7 state (including sending a CLEAR REQUEST packet to the DTE) and additionally sending a CLEAR REQUEST packet to the XDCE (via the internetworking process).

3. The use of the fast select facility with a restriction on the response prohibits the DTE from sending a CALL ACCEPT packet.

4. The DTE in the event of a call collision must discard the CALL REQUEST packet received from the DCE.

5. Table entries are defined as follows: A = action to be taken, S = the state to be entered, D = the diagnostic code to be used in packets generated as a result of this action, DISCARD indicates that the received packet is to be cleared from the XNI buffers, and INVALID indicates that the packet/state combination cannot occur.]

Table E.11-4. DTE effect on DCE reset states

DCE reset states (see Note 2)Packet Received from DTE FLOW

CONTROLREADY

d1

RESET REQUEST

by DTEd2

DCE RESETREQUEST to DTE

d3Packet with a packet type identifier shorter than 1 byte

A = ERRORS = d3D = 38

(see Note 1)


(see Note 1)

A = DISCARD

Packet with a packet type identifier that is undefined or not supported by DCE


(see Note 1)


(see Note 1)

A = DISCARD

RESTART REQUEST, or RESTART CONFIRMATION, or REGISTRATION (if supported) packet with logical channel identifier unequal to 0


(see Note 1)


(see Note 1)

A = DISCARD

DCE reset states (see Note 2)Packet Received from DTE FLOW

CONTROLREADY

d1

RESET REQUEST

by DTEd2


d3RESET REQUEST A = NORMAL

S = d2(forward)

A = DISCARD A = NORMALS = d1

(do not forward)RESET CONFIRMATION A = ERROR

S = d3D = 27

(see Note 1)


(see Note 1)

A = NORMALS = d1

(do not forward)

INTERRUPT packet See Table E.11-5 A = ERRORS = d3D = 28

(see Note 1)

A = DISCARD

INTERRUPT CONFIRMATION packet See Table E.11-5 A = ERRORS = d3D = 28

(see Note 1)

A = DISCARD

DATA or flow control packet See Table E.11-6 A = ERRORS = d3D = 28

(see Note 1)

A = DISCARD

REJECT supported but not subscribed to A = ERRORS = d3D = 37

(see Note 1)


(see Note 1)

A = DISCARD

Notes.—

1. The error procedure consists of performing the specified actions when entering the d3 state (which includes forwarding a RESET REQUEST packet to the DTE) and sending a RESET REQUEST packet to the XDCE (via the internetworking function).

2. Table entries are defined as follows: A = action to be taken, S = the state to be entered, D = the diagnostic code to be used in packets generated as a result of this action, DISCARD indicates that the received packet is to be cleared from the XNI buffers, and INVALID indicates that the packet/state combination cannot occur.

Table E.11-5. DTE effect on DCE interrupt transfer statesRESERVED

DTE/DCE interrupt transfer states (see Note 2)Packet received from DTE DTE INTERRUPT

READYi1

DTE INTERRUPT SENT

i2INTERRUPT(see Note 1)

A = NORMALS = I2

(forward)


(see Note 3)

DTE/DCE interrupt transfer states (see Note 2)Packet received from DTE DTE INTERRUPT

READYj1

DTE INTERRUPT SENT

j2INTERRUPT(see Note 1)

A = NORMALS = d3D = 43

(see Note 3)

A = ERRORS = j1

(forward)

Notes.—

1. If the packet has a format error then the error procedure applies (see Note 3).


3. The error procedure consists of performing the specified actions when entering the d3 state (which includes forwarding a RESET REQUEST packet to the DTE) and sending a RESET REQUEST packet to the XDCE (via the internetworking process).

Table E.11-7. XDCE effect on DCE restart states

DCE restart states (see Note)

Packet received from XDCEPACKET LEVEL

READYr1

DTE RESTARTREQUEST

r2

DCE RESTART REQUEST

r3CALL REQUEST See Table E.11-8 Send CLEAR

REQUEST to internetworking process

with D = 244

Send CLEAR REQUEST to

internetworking process with

D = 244CALL ACCEPT, CLEAR REQUEST, DATA, INTERRUPT, INTERRUPT CONFIRMATION, RESET REQUEST

See Table E.11-8 A = DISCARD A = DISCARD

Note.— Table entries are defined as follows: A = action to be taken, S = the state to be entered, D = the diagnostic code to be used in packets generated as a result of this action, DISCARD indicates that the received packet is to be cleared from the XNI buffers, and INVALID indicates that the packet/state combination cannot occur.

Table E.11-8. XDCE effect on DCE call setup and clearing states

DCE call setup and clearing states (see Note)Packet

receivedfrom XDCE

READY

p1

DTE CALL REQUEST

p2

DCE CALL REQUEST

p3

DATA TRANSFER

p4

CALL COLLISION

p5

DTE CLEAR

REQUEST

p6

DCE CLEAR

REQUEST to DTE

p7CALL REQUEST


INVALID INVALID INVALID INVALID INVALID INVALID

CALL ACCEPT

A = DISCARD


INVALID INVALID INVALID A = DISCARD

A = DISCARD

CLEAR REQUEST

A = DISCARD




INVALID A = DISCARD

A = DISCARD

DATA, INTERRUPT, INTERRUPT CONFIRMATION, or RESET REQUEST

A = DISCARD

INVALID INVALID See Table E.11-9

INVALID A = DISCARD

A = DISCARD

Note.— Table entries are defined as follows: A = action to be taken, S = the state to be entered, D = the diagnostic code to be used in packets generated as a result of this action, DISCARD indicates that the received packet is to be cleared from the XNI buffers, and INVALID indicates that the packet/state combination cannot occur.

Table E.11-9. XDCE effect on DCE reset states

DCE reset states (see Note)Packet Received from XDCE FLOW

CONTROLREADY

d1

DTE RESET REQUEST

d2


d3RESET REQUEST A = NORMAL

S = d3(forward)

A = NORMALS = d1

(forward)

A = DISCARD

INTERRUPT See Table E.11-10 A = DISCARD A = DISCARDINTERRUPT CONFIRMATION packet See Table E.11-10 A = DISCARD INVALIDDATA A = NORMAL A = DISCARD A = DISCARD

DCE interrupt transfer states (see Note)Packet received from XDCE DTE INTERRUPT

READYi1

DTE INTERRUPT SENT

i2

INTERRUPT CONFIRMATION INVALID A = NormalS = i1

(forward)

DCE interrupt transfer states (see Notes)

Packet received from XDCE DTE INTERRUPTREADY

j1

DCE INTERRUPT SENT

j2

INTERRUPT A = NormalS = j2

(forward)

INVALID

ADCE (GDCE) call setup and clearing States(See Notes 1, 7 and 8)

Packet receivedGNI (ANI)(See Note 2)

READY

p1

GNI (ANI) CALL

REQUEST

p2

ANI (GNI) CALL

REQUEST

p3

DATA TRANSFER

p4

GNI (ANI) CLEAR

REQUESTp6

ANI ( GNI) CLEAR

REQUEST to GNI (ANI)

p7

Format error(See Note 3)

A = ERROR(See Note 10)S = p7D = 33(See Note 9)


A = ERRORS = p7D = 33(see Notes 6 & 9)

See Table E.11-13


A = DISCARD

ADCE (GDCE) call setup and clearing States(See Notes 1, 7 and 8)

Packet receivedGNI (ANI)(See Note 2)

READY

p1

GNI (ANI) CALL

REQUEST

p2

ANI (GNI) CALL

REQUEST

p3

DATA TRANSFER

p4

GNI (ANI) CLEAR

REQUESTp6

ANI ( GNI) CLEAR


p7

CALL REQUEST A = NORMALS = p2(forward request to DCE)


Not applicable

(See Note 4)

Not applicable

(See Note 4)


A = DISCARD

CALL ACCEPT A = ERRORS = p7D = 20(See Note 10)


A= NORMALS = p4 (forward to DCE)orA = ERRORS = p7D = 42(see Note 6)



A = DISCARD

CLEAR REQUEST A = NORMALS = p6(do not forward)

A = NORMALs = p6(forward to DCE)



A = DISCARD


CLEAR CONFIRMATION

A = ERRORS = p7D = 20(See Note 10)






DATA, interrupt, flow control or reset packet




See Table E.11-13


A = DISCARD

Notes.—

1. The XDCE is not necessarily in the same state as the DTE/DCE interface.

2. All packets from the peer XNI have been checked for duplication before evaluation as represented by this table.

3. There are no detectable VDL Mode 3 format errors.

4. The ADCE assigns all channel numbers used between the ANI and GNI, hence call collisions are not possible. When a CALL REQUEST by GNI packet is received bearing a temporary channel number associated with an SVC in the p4 state, the association of the temporary to permanent channel number is broken.

5. Not applicable to the GNI.

6. The error procedure consists of performing the actions specified when entering the p7 state (including sending a CLEAR REQUEST packet to the peer XNI) and additionally sending a CLEAR REQUEST packet to the DCE (via the internetworking process).


8. The number in parentheses below an “A = NORMAL” Table entry is the paragraph number in this document that defines the actions to be taken to perform normal processing on the received packet. If no paragraph number is referenced, the normal processing is defined in the Table entry.

9. An error condition is declared and transfer to the p7 state is possible only if the ground DTE address is known unambiguously, otherwise the action is to discard the packet.

10. The error procedure consists of performing the action when entering the p7 state (including sending a CLEAR REQUEST packet to the XNI) but without sending a CLEAR REQUEST packet to the local DCE.

Table E.11-13. GNI (ANI) effect on ADCE (GDCE) reset states

ADCE (GDCE) reset states (see Notes 1, 4 and 5)

Packet Received from GNI (ANI)(See Note 2)

FLOWCONTROL

READYd1

GNI (ANI) RESET

REQUEST

d2

ADCE (GDCE) RESET


d3

RESET REQUEST A = NORMALS = d2

(forward to DCE)

A = DISCARD A = NORMALS = d1

(do not forward)

RESET CONFIRMATION A = ERRORS = d3D = 27

(see Note 3)


(see Note 3)

A = NORMALS = d1

(do not forward)

INTERRUPT See Table E.11-14 A = ERRORS = d3D = 28

(see Note 3)

A = DISCARD

INTERRUPT CONFIRMATION See Table E.11-14 A = ERRORS = d3D = 28

(see Note 3)

A = DISCARD

DATA or flow control packet See Table E.11-15 A = ERRORS = d3D = 28

(see Note 3)

A = DISCARD


Packet Received from GNI (ANI)(See Note 2)

FLOWCONTROL

READYd1

GNI (ANI) RESET

REQUEST

d2

ADCE (GDCE) RESET


d3

Format error A = ERRORS = d3D = 33

(see Note 3)


(see Note 3)

A = DISCARD

Notes.—



3. The error procedure consists of performing the specified actions when entering the d3 state (which includes forwarding a RESET REQUEST packet to the peer XNI) and sending a RESET REQUEST packet to the DCE (via the internetworking function).


5. The number in parentheses below an “A = NORMAL” table entry is the paragraph number in this document that defines the actions to be taken to perform normal processing on the received packet. If no paragraph number is referenced, the normal processing is defined in the table entry.

Table E.11-14. GNI (ANI) effect on ADCE (GDCE) interrupt transfer statesRESERVED

ADCE(GDCE) interrupt transfer states (See Notes 1, 3 and 4)

Packet received from GNI (ANI) (See Note 2)

GNI (ANI) INTERRUPTREADY

i1

GNI (ANI) INTERRUPT SENT

i2

INTERRUPT A = NORMALS = I2

(forward to DCE)


(see Note 5)



ADCE (GDCE) INTERRUPT READY

j1

ADCE (GDCE) INTERRUPT SENT

j2

INTERRUPT A = ERRORS = d3

A = NORMALS = j1



ADCE (GDCE) INTERRUPT READY

j1

ADCE (GDCE) INTERRUPT SENT

j2D = 43

(see Note 5)(forward confirmation to DCE)

Notes.—





5. The error procedure consists of performing the specified actions when entering the d3 state (which includes forwarding a RESET REQUEST packet to the peer XNI) and sending a RESET REQUEST packet to the DCE (via the internetworking function).

Table E.11-16. DCE effect on ADCE (GDCE) call setup and clearing states

ADCE (GDCE) call setup and clearing states (see Notes 1, 7 and 8)

Packet receivedfrom DCE

(See Notes 2 & 4)

READY

p1

GNI (ANI) CALL

REQUEST

p2

ADCE (GDCE) CALL

REQUEST

p3

DATA TRANSFE

R

p4

GNI (ANI) CLEAR

REQUEST

p6

ADCE (GDCE) CLEAR

REQUEST

p7

CALL REQUEST(See Note 6)


INVALID(See Note 5)

INVALID(See Note 3)

INVALID(See

Note 3)

INVALID(See Note 3)

INVALID(See

Note 3)

CALL ACCEPT(See Note 4)

A = DISCARD A = NORMALS = p4(forward)

INVALID(See Note 3)

INVALID(See

Note 3)

A = DISCARD

A = DISCARD

CLEAR REQUEST

A = DISCARD A = NORMALs = p7(forward)



A = DISCARD

DISCARD

DATA, INTERRUPT, or RESET packets(See Note 4)

A = DISCARD INVALID(See Note 3)

INVALID(See Note 3)

See Table E.11-17

A = DISCARD

A = DISCARD

Notes.—


2. This is the DTE packet received via the DCE after all DTE/DCE processing has occurred. Procedures local to the DTE/DCE interface (such as RR, RNR and REJECT if in effect), do not affect the XDCE directly. All error procedures as documented in ISO 8208 have been performed. Hence certain packets are rejected by the interface and are not represented in this table.

3. The DCE in its protocol operation with the DTE will detect this error condition, hence the erroneous packet can be said never to “reach” the XDCE; see also Note 2.

4. The channel number for the DTE/DCE need not be the same channel number used for the ADCE/GDCE; a packet from the DTE which contains a channel number is associated with an air-ground channel by means of a previously established cross reference table. If none exists then the DTE/DCE channel by definition references an air-ground channel in the p1 state.

5. The ADCE assigns all channel numbers used between the ANI and GNI; hence call collisions (denoted p5 ISO 8208) are not possible; see also Note 4.

6. A CALL REQUEST from the DTE can never be associated with an XDCE channel number that is not in the p1 state.



Table E.11-17. DCE effect on ADCE (GDCE) reset states


Packet Received from DCE FLOWCONTROL

READY

d1

GNI (ANI) RESET

REQUEST

d2

ADCE (GDCE) RESET


d3

RESET REQUEST A = NORMALS = d3

(forward)

A = NORMALS = d1

(forward)

A = DISCARD

RESET CONFIRMATION INVALID(See Note 3)

INVALID(See Note 3)

INVALID(See Note 3)

INTERRUPT See Table E.11-18 A = DISCARD Hold interrupt until VDL Mode 3 reset

complete

INTERRUPT CONFIRMATION packet See Table E.11-18 A = DISCARD INVALID(See Note 3)

DATA (See Note 2) A = NORMAL(forward)

A = DISCARD Hold interrupt until VDL Mode 3 reset

complete

Notes.—






Table E.11-18. DCE effect on ADCE (GDCE) interrupt transfer statesRESERVED

ADCE (GDCE) interrupt transfer states(See Notes 1, 4 and 5)

Packet received from DCE(See Note 2)

GNI (ANI) INTERRUPT

READYi1

GNI (ANI) INTERRUPT

SENTi2

INTERRUPT CONFIRMATION INVALID(See Note 3)

A = NormalS = i1

(forward)

ADCE (GDCE) interrupt transfer states (See Notes 1, 4 and 5)

Packet received from XDCE DTE INTERRUPTREADY

j1

DCE INTERRUPT SENT

j2

INTERRUPT A = NormalS = j2

(forward)

INVALID(See Note 3)

Notes.—






Consolidated amendments to the manual on VDL · Web viewDRAFT amendments to the manual on VHF...

Documents

Transcript of Consolidated amendments to the manual on VDL · Web viewDRAFT amendments to the manual on VHF...