chapter_3 s
Transcript of chapter_3 s
-
7/29/2019 chapter_3 s
1/32
Etisalat AcademyPage 1 of 31Anand Alexander
UMTS DOWNLINK BUDGET
"We see what we want to see unless we make a conscious effort to see
what is really there."
- Anon
Anand Alexander
-
7/29/2019 chapter_3 s
2/32
Etisalat AcademyPage 2 of 31Anand Alexander
By the end of this session, the participants will be able to:
Differentiate between GSM and UMTS link budgets
Consider all the UMTS parameter network interactions in the UMTS uplink anddownlink budget
Estimate the log normal fade margin for a % age area coverage
Calculate the uplink load factor
Illustrate the concept of uplink intra-cell noise rise and its impact on range.
Illustrate how inter-cell interference limits the effective capacity of a cell.
Objectives
-
7/29/2019 chapter_3 s
3/32
Etisalat AcademyPage 3 of 31Anand Alexander
The Downlink Link Budget Dilemma!
Downlink Range is highly dependent upon all the MobilesPositions and their individual Power Consumptions from theBase Station
UplinkRange
DownlinkRangeUplinkRange
DownlinkRange
-
7/29/2019 chapter_3 s
4/32
Etisalat AcademyPage 4 of 31Anand Alexander
UMTS Downlink Link Budget
Uplink
Range
DownlinkRange Uplink
Range
DownlinkRange
UplinkRange
DownlinkRange
UplinkRange
DownlinkRange
1 2
3 4
-
7/29/2019 chapter_3 s
5/32
Etisalat AcademyPage 5 of 31Anand Alexander
UMTS Downlink Link Budget
Max
PathLoss
Rx
Sensitivity
AntennaGain
BodyLosses
Penetr-
ationLosses
LogNormalFade
Margin
AntennaGain
DiversityGain
FeederLosses-+ + -+= Tx Power - - -
ProcessingGainInterCellInt. IntraCellInt. Eb/NoTarget
Thermal
NoisePower
NoiseFigure+ + + - +
Log
NormalFade
Margin
Soft
HandoverGain
-
Eb/NoTarget
FastFade
Margin-
BaseStation
Max.
Power
PowerConsumedIn Common
Channels
PowerConsumed
For
Handovers
PowerConsumedBy Other
Users
- - -
Other
UsersUplinks
Other
UsersLocations
-
7/29/2019 chapter_3 s
6/32
Etisalat AcademyPage 6 of 31Anand Alexander
Downlink Range is dependent upon: Usual Link Parameters (Losses, Gains, etc.)
Intracell Interference (from other user channels on base site)
Intercell Interference (from other base sites)
Available Downlink Power, which depends upon
Current Power Consumption, which depends upon The number of mobiles
Their Eb/No Downlink Targets
Their Datarates
Their Activity Factors
Their locations (distances from Base Site)
etc, etc.
UMTS Downlink Link Budget
-
7/29/2019 chapter_3 s
7/32
Etisalat AcademyPage 7 of 31Anand Alexander
Downlink Link Budget
Uplink
Range
Downlink
Range
Equal Distribution
Uplink
Range
DownlinkRange
Edge Distribution
We need to know all mobile
positions to be able to model the
downlink.
We can use some assumptions,
like assuming that all mobiles areat the cell edge, or are somehow
equally distributed.
Even with such assumptions the
Link Budget becomes a complex
task.
-
7/29/2019 chapter_3 s
8/32
Etisalat AcademyPage 8 of 31Anand Alexander
Intra cell Interference (1/2)
In the Downlink the Intracell Interference is different.
The Uplink is Asynchronous
In the Uplink the Air Interface uses the OVSF Tree to allocate different
datarates for a particular user, and therefore datarate can change on a
frame by frame basis.
In the Uplink the Air Interface uses the Scrambling codes to separate
different uses in the same Cell.Scrambling codes are of equal length and have good cross-correlation
properties (time shifted versions have low cross-correlation values).
As Uplink is Asynchronous we will receive time shifted versions of the
Scrambling Codes due to unsynchronised access, and dispersion in the
radio channel.
In the Uplink we preserve Orthogonality between users due to the propertiesof the Scrambling code. We do not have to consider Interference due to
degraded Orthogonality.
-
7/29/2019 chapter_3 s
9/32
Etisalat AcademyPage 9 of 31Anand Alexander
Intra Cell Interference (2/2)
In the Downlink the Intracell Interference is different.
The Downlink is Synchronous
In the Downlink there is only one OVSF Tree and the Air Interface uses the
OVSF Tree to separate different users in the same Cell.
In the Downlink the Air Interface uses rate adaptation and discontinuous
transmission to cater for different datarates for a particular user.In the Downlink the Air Interface uses the Scrambling Codes to separate
different Cells in the network.
OVSF codes have poor cross-correlation properties (time shifted versions
have high cross-correlation values). Orthogonality is only preserved
when not time shifted, and hence the need for the Downlink to be
Synchronised.Dispersion in the Radio Channel can cause Energy to be time shifted and
hence degrade Orthogonality between different users channels on the
Downlink.
-
7/29/2019 chapter_3 s
10/32
Etisalat AcademyPage 10 of 31Anand Alexander
DOWNLINK LOAD FACTOR
-
7/29/2019 chapter_3 s
11/32
Etisalat AcademyPage 11 of 31Anand Alexander
Load Factor
Receiver noise floor PN
K Boltzmann constant, = 1.38
T Kelvin temperature, normal temperature 290 K
W Signal bandwidth, WCDMA signal bandwidth 3.84MHz
NF: Receiver noise figure
10log(KTW) = -108dBm/3.84MHz
NF = 7dBUE typical value
NFWTKPN )**log(10
KJ/10 23
MHzdBmNFWTKPN
84.3/101)**log(10
-
7/29/2019 chapter_3 s
12/32
Etisalat AcademyPage 12 of 31Anand Alexander
Interference from users of same cell
The downlink users are identified with the mutually orthogonal OVSF codes.
In the static propagation conditions without multi-path, no mutual
interference exists.
In case of multi-path propagation, certain energy will be detected by the
RAKE receiver, and become interference signals. Define the orthogonal factor to describe this phenomenon.
PT is a total transmitting power of NodeB, which includes the dedicated
channel transmitting power and the common channel transmitting power
1 Town jjj
PI
PL
N
jCCHTPPP
1
ownI
Load Factor
-
7/29/2019 chapter_3 s
13/32
Etisalat AcademyPage 13 of 31Anand Alexander
Interference from users of adjacent cell
The transmitting signal of the adjacent cell BTS will cause interference to
the users in the current cell. Since the scrambles in use are different, such
interference is non orthogonal.
Assume the service is distributed evenly, transmitting power of all NodeBs
will be equal. In the system, there are K adjacent cell NodeBs, where path
loss from the number k NodeB to the user j is PLk,j. Hence we obtain:
K
jk
TjotherPL
PI1 ,
1
otherI
Load Factor
-
7/29/2019 chapter_3 s
14/32
Etisalat AcademyPage 14 of 31Anand Alexander
N
K
jk
T
j
T
j
NotherownTOT
PPL
PPL
P
PIII
1 ,
11
Suppose the power control is desired, we obtainSuppose the power control is desired, we obtain
jjjTOT
j
j
jvR
W
I
PL
P
NoEb1
/
ThenThen
jjTOTj
j
jjPLIv
W
RNoEbP /
AnalysisDownlink Interference Composition
Load Factor
-
7/29/2019 chapter_3 s
15/32
Etisalat AcademyPage 15 of 31Anand Alexander
jN
K
jk
j
TTj
N
j
j
jCCH
N
K
jk
T
j
T
j
N
jj
j
jCCH
N
jjTOTj
j
jCCHT
PLPPL
PLPPv
W
RNoEbP
PPL
PPL
PPLv
W
RNoEbP
PLIvW
RNoEbPP
1 ,1
1 ,1
1
1/
11/
/
N
jCCHTPPP
1
Because
Downlink Interference Analysis
Then
Load Factor
-
7/29/2019 chapter_3 s
16/32
Etisalat AcademyPage 16 of 31Anand Alexander
N
j
j
jjj
N
jj
j
jNCCH
T
vW
R
NoEbi
PLvW
RNoEbPP
P
1
1
/11
/
K
jk
jj
PLPLi
1 ,
Resolve PT to obtain
where i j is the adjacent cell interference factor of the user, defined as:
Downlink Interference Analysis
Load Factor
-
7/29/2019 chapter_3 s
17/32
Etisalat AcademyPage 17 of 31Anand Alexander
N
j
j
jjjDLv
W
RNoEbi
1
/1
Downlink Interference Analysis
According to the analysis, we can define the downlink load factor:
As different from the theoretic calculation of uplink capacity, and in the
downlink capacity formula are variable related to user position. Namely, the
downlink capacity is related to the spatial distribution of the users, and can
only be determined through system emulation.
Load Factor
-
7/29/2019 chapter_3 s
18/32
Etisalat AcademyPage 18 of 31Anand Alexander
Load Factor
M
j
j
j
jo
b
DLW
R
N
Ei 1
Average Downlink Load Factor is presented, based upon usingaverage values for the Orthogonality factors, j, and Other Cell to
Own Cell Powers, ij. This results in a modified equation as:
If all Musers in the Cell were using the same type of service, then
Eb/No, Activity Rate, and Bit Rate would be the same.
In this case we can state that the Average Downlink Load Factor,
DL can be expressed as:
W
R
N
EMi
o
b
DL1
-
7/29/2019 chapter_3 s
19/32
Etisalat AcademyPage 19 of 31Anand Alexander
Emulation Result
When Node B Tx is 43dBm(20W), the maximum numberof users is approx. 114
To ensure system stability, themean Tx power of Node B
should not be more than 80%of the maximum Tx power, 42dBm. This way, the supportednumber of users is 110
-
7/29/2019 chapter_3 s
20/32
Etisalat AcademyPage 20 of 31Anand Alexander
Coverage/Capacity
Downlink Noise Rise as a Function of Downlink Data Throughput and
i for avg = 0.6 (ITU Vehicular A Channel)
0
2
4
6
8
10
12
14
16
18
20
0 1000 2000 3000 4000 5000 6000
Throughput (kbps)
NoiseRise(dB)
10%
25%
50%
75%
90%
Average i
Downlink Noise Rise as a Function of Downl ink Data Throughput and
i for avg = 0.9 (ITU Pedestrian A Channel)
0.00
2.00
4.00
6.00
8.00
10.00
12.00
14.00
16.00
18.00
20.00
0 1000 2000 3000 4000 5000 6000
Throughput (kbps)
No
iseRise(dB)
10%
25%
50%
75%
90%
Average i
Downlink Noise Rise as a function ofdata throughput.
Assumes:
Eb/No = 5.5dB
User Average i= 10% to 90%
LCD144 Users
User Average a = 0.6 and 0.9 50% Cell Load
Intercell Interference (from iavg) andIntracell Orthogonality (from aavg)limits Pole Capacity
Intracell Interference (additionalthroughput) limits range
-
7/29/2019 chapter_3 s
21/32
Etisalat AcademyPage 21 of 31Anand Alexander
Coverage/Capacity
UMTS Dow nlink Range as a function of Capacity and Average User i
for avg = 0.6 (ITU Vehicular Channel A)
0
1
2
3
4
5
6
0 1000 2000 3000 4000 5000 6000
Throughput (kbps)
Range(km)
10%
25%
50%
75%
90%
UMTS Down link Range as a function of Capacity and Average User i
for avg = 0.9 (ITU Pedestrian Channel A)
0
1
2
3
4
5
6
0 1000 2000 3000 4000 5000 6000
Throughput (kbps)
R
ange(km)
10%
25%
50%
75%
90%
Average i
Average i
DownlinkRange as a function of datathroughput.
Assumes:
BS Power = 20W
Eb/No = 5.5dB
User Average i = 10% to 90%
LCD144 Users User Average = 0.6 and 0.9
50% Cell Load
Intercell Interference (from iavg) andIntracell Orthogonality (from aavg)limits Pole Capacity
Intracell Interference (additional
throughput) limits range
-
7/29/2019 chapter_3 s
22/32
Etisalat AcademyPage 22 of 31Anand Alexander
Coverage/Capacity
Uplink and Downlink range as a function of
capacity, or throughput, are shown together.
LCD144 Services
Uplink:
Eb/No = 1.5dB
i = 0.65
Downlink:
Eb/No = 5.5dB
iavg = 0.8
aavg = 0.6
UMTS Uplink and Downlink Range as a function of Uplink and Downlink
Capacity
0.00
0.50
1.00
1.50
2.00
2.50
3.00
0 500 1000 1500
Load (kbps)
CellRadius(km)
UMTS Uplink and Downlink Range as a function of Uplink and Downlink
Capacity
0.00
0.50
1.00
1.50
2.00
2.50
3.00
0 500 1000 1500
Load (kbps)
CellRadius(km)
UMTS Uplink and Downlink Range as a function of Uplink and Downlink
Capacity
0.00
0.50
1.00
1.50
2.00
2.50
3.00
0 500 1000 1500
Load (kbps)
CellRadius(km)
UMTS Uplink and Downlink Range as a function of Uplink and Downlink
Capacity
0.00
0.50
1.00
1.50
2.00
2.50
3.00
0 500 1000 1500
Load (kbps)
CellRadius(km)
Downlink Coverage/Capacity values forcombinations of User Positions, and Cell Loading
I.e. due to User Movement and Loading
Uplink Coverage/Capacity values forcombinations of Cell Loading.
Downlink
Uplink
-
7/29/2019 chapter_3 s
23/32
Etisalat AcademyPage 23 of 31Anand Alexander
Coverage/Capacity
In WCDMA/UMTS there exists awhole range of possible Capacity
and Coverage combinations, based
upon Service Mixes, user speeds,
Interference Geometry, User
Positions, Channel Multipath, etc,
etc.
In contrast with GSM there existsessentially one Capacity/Coverage
point, and is not dependent upon
user locations, Service mix, user
speeds, etc.
-
7/29/2019 chapter_3 s
24/32
Etisalat AcademyPage 24 of 31Anand Alexander
Impact of Soft(er) Handover
Handover Area where DownlinkPilot Power is within xdB of
each other and within Range
Large Handover Area = GoodResilience for MS at cell edge, given
that Cell can breathe, but lowercapacity
Small Handover Area = PoorResilience for MS at cell edge, giventhat Cell can breathe, but higher
capacity
75%Load Range
75%Load Range
-
7/29/2019 chapter_3 s
25/32
Etisalat AcademyPage 25 of 31Anand Alexander
Impact of Soft(er) Handover
-
7/29/2019 chapter_3 s
26/32
Etisalat AcademyPage 26 of 31Anand Alexander
LINK BUDGET ANALYSIS
We shall look at the impact on the Network Design of:
Antenna Downtilt
Antenna Sectorisation
Mast Head Amplifiers
-
7/29/2019 chapter_3 s
27/32
Etisalat AcademyPage 27 of 31Anand Alexander
Network Simulation
Parameters used in simulation.
13.5km2 of Tokyo
10 Sites, 50m Height
20W Base Station Power
15dB Penetration Losses = 12dB
Channel Profile = ITUVehicular 3km/h
Average UserOrthogonality,
avg
= 0.5
Soft Handover AdditionWindow = 4dB
-
7/29/2019 chapter_3 s
28/32
Etisalat AcademyPage 28 of 31Anand Alexander
Analysis
-
7/29/2019 chapter_3 s
29/32
Etisalat AcademyPage 29 of 31Anand Alexander
Antenna Tilts
With Antenna Downtilt, one would expectthe Intercell Interference to be better
contained, at the expense of reducing
Coverage Quality.
The Table shows the results obtained four
types of Antenna.
The graph homes in on the tri-sectoredantenna.
i decreases with Tilt Angle
No. of Users increase with Tilt
Angle
Coverage increases then decreases
with Antenna Tilt
Antenna Tilt
Other/Own Cell
Interference
Ratio, i
Served
Users
Soft
Handover
Overhead8kbps 64kbps 144kbps
0o
0.79 239 28% 70% 32% 40%
0o 0.88 575 40% 86% 59% 62%
4o
0.75 624 39% 91% 71% 72%
7o
0.59 697 36% 92% 76% 76%
10o 0.37 856 30% 90% 75% 74%
14o
0.38 787 32% 81% 62% 61%
0o 1.09 604 41% 92% 70% 71%
4o
0.94 707 30% 95% 81% 81%
7o
0.72 833 26% 96% 84% 83%
10o 0.47 959 21% 94% 82% 81%
14o
0.50 886 26% 86% 69% 68%
0o 1.15 880 48% 93% 76% 76%
4o 1.03 946 49% 96% 83% 83%
7o
0.88 1037 45% 96% 85% 84%
10o 0.73 1054 41% 95% 83% 82%
14o
0.58 930 33% 86% 70% 69%
UL Coverage Probability
3-Sectored, 65o
4-Sectored, 65o
6-Sectored, 65o
Omni
Uplink i and Cell capacity as a Function of Antenna Tilt
for 3-Sectored 65o
antennas
0
100
200
300400
500
600
700
800
900
0 2 4 6 8 10 12 14
Antenna Tilt
Numbe
rofUsers
(Capacity)
0.00
0.10
0.20
0.30
0.40
0.50
0.60
0.70
0.80
0.90
1.00
OtherCe
ll/OwnCell
Interfe
rence,i
Served Users
Other/Own Cell Interference Ratio, i
-
7/29/2019 chapter_3 s
30/32
Etisalat AcademyPage 30 of 31Anand Alexander
Antenna Beamwidth
MHA in use, No Downtilt, MS Tx
Power = 24dBm Max.
Higher Sectorisation, More Capacity
per site achieved.
Narrower Beamwidth results in lower
Other Cell to Own Cell Interference.
Narrow Beamwidth results in moreCapacity and Reduction in SoftHandover Overhead
Coverage Probability has an Optimum
value.
Antenna
Beamwidth
Other/Own Cell
Interference
Ratio, i
Served
Users
Soft
Handover
Overhead
8kbps 64kbps 144kbps
360o
0.79 240 28% 70% 32% 40%
120o
1.33 441 39% 85% 50% 59%
90o
1.19 461 35% 87% 55% 62%
65o 0.88 575 34% 86% 59% 62%
120o 1.72 489 54% 90% 62% 68%
90o 1.49 510 51% 92% 67% 72%
65o
1.09 604 41% 92% 70% 71%
33o
0.92 691 40% 88% 65% 64%
120o
2.18 593 64% 95% 75% 79%90
o1.97 627 59% 96% 80% 82%
65o 1.43 758 55% 96% 80% 81%
33o 1.15 880 48% 93% 76% 76%
6-Sectored
UL Coverage Probability
Omni
3-Sectored
4-Sectored
Uplink Coverage probability and Users Served as a Function
of Antenna Sectorisation and Beamwidth
84%
86%
88%
90%
92%
94%
96%
98%
0 100 200 300 400 500 600 700 800 900 1 000
No. of Users
Coverage
Probability(8Kbps 3-Sectored 120deg
3-Sectored 90deg3-Sectored 65deg4-Sectored 120deg4-Sectored 90deg4-Sectored 65deg6-Sectored 120deg6-Sectored 90deg6-Sectored 65deg
6-Sectored 33deg4-Sectored 33deg
3-sectors
4-sectors
6-sectors
Uplink Coverage probability and Users Served as a Function
of Antenna Sectorisation and Beamwidth
84%
86%
88%
90%
92%
94%
96%
98%
0 100 200 300 400 500 600 700 800 900 1 000
No. of Users
Coverage
Probability(8Kbps 3-Sectored 120deg
3-Sectored 90deg3-Sectored 65deg4-Sectored 120deg4-Sectored 90deg4-Sectored 65deg6-Sectored 120deg6-Sectored 90deg6-Sectored 65deg
6-Sectored 33deg4-Sectored 33deg
3-sectors
4-sectors
6-sectors
90o
120o65o
120o
120o
90o
65o
65o90o
33o
33o
L N i M H d A lifi (MHA) U li k
-
7/29/2019 chapter_3 s
31/32
Etisalat AcademyPage 31 of 31Anand Alexander
Low Noise Mast Head Amplifier (MHA) on Uplink
MHA
Other/Own CellInterference
Ratio, i
ServedUsers in
UL
Servedusers in
DL
8kbps 64kbps 144kbps
no MHA 0.60 1038 807 93% 78% 78%
with MHA 0.61 1064 746 95% 82% 82%
no MHA 0.73 1089 884 96% 86% 85%
with MHA 0.73 1107 846 98% 89% 89%
no MHA 0.88 1124 1052 97% 87% 86%
with MHA 0.90 1132 1021 98% 90% 90%
UL Coverage Probability
3-Sectored, 65o
4-Sectored, 65
o
6-Sectored, 65o
Antenna Tilt = 7o
; MS Power = 27dBm Increase in Number of UL users with MHA
Decrease in Number of DL users with MHA
Increase in UL Coverage Probability with MHA
-
7/29/2019 chapter_3 s
32/32
Question and Answers
Discussion