Doc.: IEEE 802.22-11/0022r0 Submission February 2011 Gerald Chouinard, CRCSlide 1 Best static tone...
-
Upload
jonas-ryan -
Category
Documents
-
view
214 -
download
0
description
Transcript of Doc.: IEEE 802.22-11/0022r0 Submission February 2011 Gerald Chouinard, CRCSlide 1 Best static tone...
Gerald Chouinard, CRCSlide 1
doc.: IEEE 802.22-11/0022r0
Submission
February 2011
Best static tone locations for extracting upstream channel impulse response and fine ranging
Name Company Address Phone email Gerald Chouinard
Communications Research Centre, Canada
3701 Carling Ave. Ottawa, Ontario Canada K2H 8S2
(613) 998-2500
Authors:
Notice: This document has been prepared to assist IEEE 802.22. It is offered as a basis for discussion and is not binding on the contributing individual(s) or organization(s). The material in this document is subject to change in form and content after further study. The contributor(s) reserve(s) the right to add, amend or withdraw material contained herein.
Release: The contributor grants a free, irrevocable license to the IEEE to incorporate material contained in this contribution, and any modifications thereof, in the creation of an IEEE Standards publication; to copyright in the IEEE’s name any IEEE Standards publication even though it may include portions of this contribution; and at the IEEE’s sole discretion to permit others to reproduce in whole or in part the resulting IEEE Standards publication. The contributor also acknowledges and accepts that this contribution may be made public by IEEE 802.11.
Patent Policy and Procedures: The contributor is familiar with the IEEE 802 Patent Policy and Procedures <http://standards.ieee.org/guides/bylaws/sb-bylaws.pdf>, including the statement "IEEE standards may include the known use of patent(s), including patent applications, provided the IEEE receives assurance from the patent holder or applicant with respect to patents essential for compliance with both mandatory and optional portions of the standard." Early disclosure to the Working Group of patent information that might be relevant to the standard is essential to reduce the possibility for delays in the development process and increase the likelihood that the draft publication will be approved for publication. Please notify the Chair Carl R. Stevenson <[email protected]> as early as possible, in written or electronic form, if patented technology (or technology under patent application) might be incorporated into a draft standard being developed within the IEEE 802.11 Working Group. If you have questions, contact the IEEE Patent Committee Administrator at <[email protected]>.
AbstractThis contribution presents the results of a parametric study conducted at CRC in order to identify the performance of a number of upstream static subcarriers sets in an attempt to maximize the dynamic range of echo detection capability while trying to maximize the echo delay range before aliasing starts to appear and maximize the bandwidth for a minimum number of static subcarriers in the P802.22 system upstream. This is in response to comment #209 to the P802.22 D1.0.
Gerald Chouinard, CRCSlide 2
doc.: IEEE 802.22-11/0022r0
Submission
February 2011
Outline1. Extraction of the Complex Channel Impulse Response2. Generation of the Prototype Function
(i.e., High-resolution Complex Channel Impulse Response)
3. Channel deconvolution process4. Some subcarrier patterns for channel deconvolution
• Examples of perfect Channel Impulse Responses• Examples of Prototype Functions• Resulting Channel Echo Functions for various subcarrier patterns
5. Summary of the exercise6. Summary of the results7. Observations on the results8. Proposal
Gerald Chouinard, CRCSlide 3
doc.: IEEE 802.22-11/0022r0
Submission
February 2011
Outline1. Extraction of the Complex Channel Impulse Response2. Generation of the Prototype Function
(i.e., High-resolution Complex Channel Impulse Response)
3. Channel deconvolution process4. Some subcarrier patterns for channel deconvolution
• Examples of perfect Channel Impulse Responses• Examples of Prototype Functions• Resulting Channel Echo Functions for various subcarrier patterns
5. Summary of the exercise6. Summary of the results7. Observations on the results8. Proposal
Gerald Chouinard, CRCSlide 4
doc.: IEEE 802.22-11/0022r0
Submission
February 2011
802.22 OFDM Subcarrier Set
0-1-840 +840+1Subcarrier index
Am
plitu
de
Gerald Chouinard, CRCSlide 5
doc.: IEEE 802.22-11/0022r0
Submission
February 2011
Extraction of the complex Channel Impulse Response from the LTS preamble on the downstream
Syncadvance IQ Vector
LTS
Frequency
...
IDFT
Time
QI
Time
Cyclic prefix
QI
QI
QI
Time
2048 samples
QI
DFT
LTS distortedby channel
Frequency
...
QI
QI
τ1
Dirac distortedby channel
Frequency
...
Carrier phase reversalbased on the LTS coding
QI
QI
QI
Convolution with channelimpulse response
QI
IDFT
Complex channel impulse response relativeto the receiver synchronization time
QI
Sampling time
Imaginary
Rea
l
Gerald Chouinard, CRCSlide 6
doc.: IEEE 802.22-11/0022r0
Submission
February 2011
Channel B
Gerald Chouinard, CRCSlide 7
doc.: IEEE 802.22-11/0022r0
Submission
February 2011
Example LTS generated multipath response
20 40 60 80 100 120 140 160 180 200-1
-0.8
-0.6
-0.4
-0.2
0
0.2
0.4
0.6
0.8
1
Time samples
Am
plitu
de
RealAbs
(1 sample = 145.86 ns)
SNR= 0 dB
Channel profile model B
Gerald Chouinard, CRCSlide 8
doc.: IEEE 802.22-11/0022r0
Submission
February 2011
Example LTS generated multipath response
20 40 60 80 100 120 140 160 180 200-1
-0.8
-0.6
-0.4
-0.2
0
0.2
0.4
0.6
0.8
1
Time samples
Am
plitu
de
RealAbs
(1 sample = 145.86 ns)
SNR= 0 dB
Channel profile model B
Gerald Chouinard, CRCSlide 9
doc.: IEEE 802.22-11/0022r0
Submission
February 2011
Example LTS generated multipath response
5 10 15 20 25 30 35 40 45 50 55 60-1
-0.8
-0.6
-0.4
-0.2
0
0.2
0.4
0.6
0.8
1
Time samples
Am
plitu
de
RealAbs
(1 sample = 145.86 ns)
SNR= 0 dB
Channel profile model B
Gerald Chouinard, CRCSlide 10
doc.: IEEE 802.22-11/0022r0
Submission
February 2011
Outline1. Extraction of the Complex Channel Impulse Response2. Generation of the Prototype Function
(i.e., High-resolution Complex Channel Impulse Response)
3. Channel deconvolution process4. Some subcarrier patterns for channel deconvolution
• Examples of perfect Channel Impulse Responses• Examples of Prototype Functions• Resulting Channel Echo Functions for various subcarrier patterns
5. Summary of the exercise6. Summary of the results7. Observations on the results8. Proposal
Gerald Chouinard, CRCSlide 11
doc.: IEEE 802.22-11/0022r0
Submission
February 2011
Construction of the “high-resolution” Prototype Function
60007000
80009000
10000
-1
-0.5
0
0.5
1-1
-0.5
0
0.5
1
Precise time sample
LTS prototype function
Imaginary
Rea
l
146 ns
Stimulus
=146/180= 0.81 ns
0-1-840 +840+1Subcarrier index
Am
plitu
de
IDFT [ *e-jwt]
Gerald Chouinard, CRCSlide 12
doc.: IEEE 802.22-11/0022r0
Submission
February 2011
Construction of the “high-resolution” Prototype Function
60007000
80009000
10000
-1
-0.5
0
0.5
1-1
-0.5
0
0.5
1
Precise time sample
LTS prototype function
Imaginary
Rea
l
146 ns
Stimulus
=146/180= 0.81 ns
0-1-840 +840+1Subcarrier index
Am
plitu
de
IDFT [ *e-jwt]
Gerald Chouinard, CRCSlide 13
doc.: IEEE 802.22-11/0022r0
Submission
February 2011
Construction of the “high-resolution” Prototype Function
60007000
80009000
10000
-1
-0.5
0
0.5
1-1
-0.5
0
0.5
1
Precise time sample
LTS prototype function
Imaginary
Rea
l
=146/180= 0.81 ns146 ns
Stimuli
0-1-840 +840+1Subcarrier index
Am
plitu
de
IDFT [ *e-jwt]
Gerald Chouinard, CRCSlide 14
doc.: IEEE 802.22-11/0022r0
Submission
February 2011
Construction of the “high-resolution” Prototype Function(Real view)
6000 6500 7000 7500 8000 8500 9000 9500 10000-1
-0.8
-0.6
-0.4
-0.2
0
0.2
0.4
0.6
0.8
1LTS prototype function
Precise time sample
Rea
l
=146/180= 0.81 ns
0-1-840 +840+1Subcarrier index
Am
plitu
de
IDFT [ ]
Gerald Chouinard, CRCSlide 15
doc.: IEEE 802.22-11/0022r0
Submission
February 2011
Construction of the “high-resolution” Prototype Function(Complex view)
60007000
80009000
10000
-1
-0.5
0
0.5
1-1
-0.5
0
0.5
1
Precise time sample
LTS prototype function
Imaginary
Rea
l
=146/180= 0.8 ns
Gerald Chouinard, CRCSlide 16
doc.: IEEE 802.22-11/0022r0
Submission
February 2011
Outline1. Extraction of the Complex Channel Impulse Response2. Generation of the Prototype Function
(i.e., High-resolution Complex Channel Impulse Response)
3. Channel deconvolution process4. Some subcarrier patterns for channel deconvolution
• Examples of perfect Channel Impulse Responses• Examples of Prototype Functions• Resulting Channel Echo Functions for various subcarrier patterns
5. Summary of the exercise6. Summary of the results7. Observations on the results8. Proposal
Gerald Chouinard, CRCSlide 17
doc.: IEEE 802.22-11/0022r0
Submission
February 2011
Channel deconvolution process
Complexcorrelation
Channel impulse responserelative to the sampling time
τ 1
τ 2τ 3Amplitude1 Delay1Amplitude2 Delay2Amplitude3 Delay3Amplitude4 Delay4 etc...
2048 I&Q samples at samplingperiod (i.e., every 145.86 ns)
High resolution bandlimited impulse response
(e.g., every 0.81 ns)
QI
2048 x 180 I&Q samplesat every 0.81 nsQI
-1
01
Precise time sampleImaginary
Rea
l
-1
0
1
I
Q
Channel impulse responserelative to the sampling time
Sampling times
ImaginaryR
eal
Gerald Chouinard, CRCSlide 18
doc.: IEEE 802.22-11/0022r0
Submission
February 2011
Example of correlation output waveform
Precise time samples
Cor
rela
tion
Out
put A
mpl
itude
0.5 1 1.5 2 2.5 3 3.5 x 104
-1
-0.8
-0.6
-0.4
-0.2
0
0.2
0.4
0.6
0.8
1Real
Samples
(1 sample = 0.81 ns)30 µs
SNR= 0 dB
Channel profile model B
Gerald Chouinard, CRCSlide 19
doc.: IEEE 802.22-11/0022r0
Submission
February 2011
Example of correlation output waveform
1000 2000 3000 4000 5000 6000 7000 8000 9000 10000-1
-0.8
-0.6
-0.4
-0.2
0
0.2
0.4
0.6
0.8
1
Precise time samples
Cor
rrel
atio
n O
utpu
t Am
plitu
de
RealSamples
(1 sample = 0.81 ns)30 µs
SNR= 0 dB
Channel profile model B
Gerald Chouinard, CRCSlide 20
doc.: IEEE 802.22-11/0022r0
Submission
February 2011
7500 8000 8500 9000 9500
0
0.05
0.1
0.15
0.2
0.25
0.3
0.35
0.4
0.45
Correlation Response
Precise time samples
Rea
l and
Imag
inar
y A
mpl
itude
s
Imag Real Samples
Example of correlation output waveform
(1 sample = 0.81 ns)
High resolution (Real) Low resolution (Abs) Samples
Sampling period interpolationInterpolation
within the sampling period
Gerald Chouinard, CRCSlide 21
doc.: IEEE 802.22-11/0022r0
Submission
February 2011
8763 8764 8765 8766 8767 8768 8769 8770 8771
0.3932
0.3934
0.3936
0.3938
0.394
0.3942
0.3944
0.3946
0.3948
0.395
Correlation Response
Precise time samples
Rea
l and
Imag
inar
y A
mpl
itude
s
Imag Real Samples
Example of correlation output waveform
(1 sample = 0.81 ns)
3rd echo delay relative to RX sync = 8767 micro-samples
Noise: some 30 dB belowat SNR= 6 dB
High resolution (Real)
= 48 +127/180 samples
Channel profile model B
= 7 usec relative to Rx sync
Gerald Chouinard, CRCSlide 22
doc.: IEEE 802.22-11/0022r0
Submission
February 2011
Outline1. Extraction of the Complex Channel Impulse Response2. Generation of the Prototype Function
(i.e., High-resolution Complex Channel Impulse Response)
3. Channel deconvolution process4. Some subcarrier patterns for channel deconvolution
• Examples of perfect Channel Impulse Responses• Examples of Prototype Functions• Resulting Channel Echo Functions for various subcarrier patterns
5. Summary of the exercise6. Summary of the results7. Observations on the results8. Proposal
Gerald Chouinard, CRCSlide 23
doc.: IEEE 802.22-11/0022r0
Submission
February 2011
Subcarrier patterns for channel deconvolutionDownstream:• Frame preamble
– Long training sequence: 840 subcarriers (one every two subcarriers)– Maximum echo delay range without aliasing: 149.4 µsec
Upstream:• CDMA Ranging burst
– Extensive search was conducted to find the optimum set of unevenly spread 56 subcarriers that would allow a multipath detection range of 30 µsec (echoes appearing beyond this range on the right would create aliasing and start to appear on the left of the range as ‘close-in’ pre-echoes)
– The resulting perfect channel impulse response was found to be limited to 20 dB sidelobe rejection (see slide 30) which is insufficient (re.: 22-10-0178r1). The corresponding Prototype Function (see slide 35) reduced this rejection to 17 dB.
– It was decided to expand the search to 2, 3 and 6 sub-channels (56, 84 and 168 subcarriers) with evenly spread subcarriers to achieve better performance.
Gerald Chouinard, CRCSlide 24
doc.: IEEE 802.22-11/0022r0
Submission
February 2011
Outline1. Extraction of the Complex Channel Impulse Response2. Generation of the Prototype Function
(i.e., High-resolution Complex Channel Impulse Response)
3. Channel deconvolution process4. Some subcarrier patterns for channel deconvolution
• Examples of perfect Channel Impulse Responses• Examples of Prototype Functions• Resulting Channel Echo Functions for various subcarrier patterns
5. Summary of the exercise6. Summary of the results7. Observations on the results8. Proposal
Gerald Chouinard, CRCSlide 25
doc.: IEEE 802.22-11/0022r0
Submission
February 2011
Total perfect channel impulse response(1680 subcarriers over 6 MHz)
200 400 600 800 1000 1200 1400 1600 1800 2000-70
-60
-50
-40
-30
-20
-10
0CIR for 1680 subcarriers distributed every subcarrier over the 5.625 MHz bandwidth
Time samples (1 sample = 145.83 ns, Entire span = 298.655 usec)
Rel
ativ
e A
mpl
itude
(dB
)
Sub-sampled function
300 µs
Gerald Chouinard, CRCSlide 26
doc.: IEEE 802.22-11/0022r0
Submission
February 2011
Downstream LTS perfect channel impulse response(840 subcarriers over 6 MHz)
200 400 600 800 1000 1200 1400 1600 1800 2000-70
-60
-50
-40
-30
-20
-10
0CIR for 840 subcarriers distributed every 2 subcarrier over the 5.625 MHz bandwidth
Time samples (1 sample = 145.83 ns, Entire span = 298.655 usec)
Rel
ativ
e A
mpl
itude
(dB
)
Sub-sampled functionEntire function
300 µs
Gerald Chouinard, CRCSlide 27
doc.: IEEE 802.22-11/0022r0
Submission
February 2011
Downstream LTS channel impulse response(840 subcarriers, close-in sidelobes)
2 4 6 8 10 12 14 16 18 20-70
-60
-50
-40
-30
-20
-10
0CIR for 840 subcarriers distributed every 2 subcarrier over the 5.625 MHz bandwidth
Time samples (1 sample = 145.83 ns, Entire span = 2.917 usec)
Rel
ativ
e A
mpl
itude
(dB
)
Sub-sampled functionEntire function
3 µs
Note the limited rejection of the first
sidelobes. This limitation will be removed by the
complex correlation.
Gerald Chouinard, CRCSlide 28
doc.: IEEE 802.22-11/0022r0
Submission
February 2011
Upstream channel impulse response (168 subcarriers)
200 400 600 800 1000 1200 1400 1600 1800 2000-70
-60
-50
-40
-30
-20
-10
0CIR for 168 subcarriers distributed every 10 subcarriers over the 5.625 MHz bandwidth
Time samples (1 sample = 145.83 ns, Entire span = 298.655 usec)
Rel
ativ
e A
mpl
itude
(dB
)
Sub-sampled functionEntire function
300 µs
30 µs
Gerald Chouinard, CRCSlide 29
doc.: IEEE 802.22-11/0022r0
Submission
February 2011
Upstream channel impulse response(168 subcarriers, close-in sidelobes)
2 4 6 8 10 12 14 16 18 20-70
-60
-50
-40
-30
-20
-10
0CIR for 168 subcarriers distributed every 10 subcarriers over the 5.625 MHz bandwidth
Time samples (1 sample = 145.83 ns, Entire span = 2.917 usec)
Rel
ativ
e A
mpl
itude
(dB
)
Sub-sampled functionEntire function
3 µs
Note the limited rejection of the first
sidelobes. This limitation will be removed by the
complex correlation.
Gerald Chouinard, CRCSlide 30
doc.: IEEE 802.22-11/0022r0
Submission
February 2011
50 100 150 200 250 300 350 400 450 500-60
-50
-40
-30
-20
-10
0[28/42]: 56-carriers prototype function spread over +/-1.5 MHz
Time samples
Rel
ativ
e A
mpl
itude
(dB
)
[1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1][1 1 1 1 1 1 1 1 1 1 1 0 0 0 0 1 1 1 1 0 1 1 1 0 1 0 1 0 1 1 0 0 1 1 0 0 1 0 1 1 0 1]
Upstream channel impulse response(56 subcarriers, irregular spacing)
(1 sample = 58.3 ns)
168 -
carr
i er f
unct
ion
56- c
arrie
r fu n
c tio
n
15 µs
CIR for 56 subcarriers irregularly distributed over 2.814 MHz
Gerald Chouinard, CRCSlide 31
doc.: IEEE 802.22-11/0022r0
Submission
February 2011
Outline1. Extraction of the Complex Channel Impulse Response2. Generation of the Prototype Function
(i.e., High-resolution Complex Channel Impulse Response)
3. Channel deconvolution process4. Some subcarrier patterns for channel deconvolution
• Examples of perfect Channel Impulse Responses• Examples of Prototype Functions• Resulting Channel Echo Functions for various subcarrier patterns
5. Summary of the exercise6. Summary of the results7. Observations on the results8. Proposal
Gerald Chouinard, CRCSlide 32
doc.: IEEE 802.22-11/0022r0
Submission
February 2011
0.2 0.4 0.6 0.8 1 1.2 1.4 1.6 1.8 2 2.2 2.4
x 104
-45
-40
-35
-30
-25
-20
-15
-10
-5
0840-subcarrier over 6 MHz prototype function
Precise time sample
Rel
ativ
e am
plitu
de (d
B)
0.5 1 1.5 2 2.5 3 3.5
x 104
-45
-40
-35
-30
-25
-20
-15
-10
-5
0840-subcarrier over 6 MHz prototype function
Precise time sample
Rel
ativ
e am
plitu
de (d
B)
840-subcarrier prototype function in 6 MHz
9.7 µs0 µs-9.7 µs
Gerald Chouinard, CRCSlide 33
doc.: IEEE 802.22-11/0022r0
Submission
February 2011
0.2 0.4 0.6 0.8 1 1.2 1.4 1.6 1.8 2 2.2 2.4
x 104
-45
-40
-35
-30
-25
-20
-15
-10
-5
0168-subcarrier over 6 MHz prototype function
Precise time sample
Rel
ativ
e am
plitu
de (d
B)
168-subcarrier prototype function in 6 MHz
9.7 µs0 µs-9.7 µs
Gerald Chouinard, CRCSlide 34
doc.: IEEE 802.22-11/0022r0
Submission
February 2011
0.2 0.4 0.6 0.8 1 1.2 1.4 1.6 1.8 2 2.2 2.4
x 104
-45
-40
-35
-30
-25
-20
-15
-10
-5
084-subcarrier over 3 MHz prototype function
Precise time sample
Rel
ativ
e am
plitu
de (d
B)
84-subcarrier prototype function in 3 MHz
9.7 µs0 µs-9.7 µs
Gerald Chouinard, CRCSlide 35
doc.: IEEE 802.22-11/0022r0
Submission
February 2011
0.2 0.4 0.6 0.8 1 1.2 1.4 1.6 1.8 2 2.2 2.4
x 104
-45
-40
-35
-30
-25
-20
-15
-10
-5
056-subcarrier over 3 MHz prototype function
Precise time sample
Rel
ativ
e am
plitu
de (d
B)
56-subcarrier prototype function in 3 MHz(uneven spreading)
9.7 µs0 µs-9.7 µs
Gerald Chouinard, CRCSlide 36
doc.: IEEE 802.22-11/0022r0
Submission
February 2011
Outline1. Extraction of the Complex Channel Impulse Response2. Generation of the Prototype Function
(i.e., High-resolution Complex Channel Impulse Response)
3. Channel deconvolution process4. Some subcarrier patterns for channel deconvolution
• Examples of perfect Channel Impulse Responses• Examples of Prototype Functions• Resulting Channel Echo Functions for various subcarrier patterns
5. Summary of the exercise6. Summary of the results7. Observations on the results8. Proposal
Gerald Chouinard, CRCSlide 37
doc.: IEEE 802.22-11/0022r0
Submission
February 2011
Channel echo function for 840 subcarriersspread over 5.63 MHz (SNR = 6 dB)
1000 2000 3000 4000 5000 6000 7000 8000 9000 10000-0.4
-0.2
0
0.2
0.4
0.6
0.8
1Correlation Response for 840 subcarriers in 6 MHz
Precise time samples
Am
plitu
des
HR Chan. FunctionCCIR Abs samples
1000 2000 3000 4000 5000 6000 7000 8000 9000 10000-50
-45
-40
-35
-30
-25
-20
-15
-10
-5
0Correlation Response for 840 subcarriers in 6 MHz
Precise time samples
Am
plitu
des
(dB
)
8 µs 8 µs
Gerald Chouinard, CRCSlide 38
doc.: IEEE 802.22-11/0022r0
Submission
February 2011
Channel echo function for 168 subcarriers spreadover 5.63 MHz (SNR = 6 dB)
1000 2000 3000 4000 5000 6000 7000 8000 9000 10000-0.4
-0.2
0
0.2
0.4
0.6
0.8
1Correlation Response for 168 subcarriers in 6 MHz
Precise time samples
Am
plitu
des
HR Chan. FunctionCCIR Abs samples
1000 2000 3000 4000 5000 6000 7000 8000 9000 10000-50
-45
-40
-35
-30
-25
-20
-15
-10
-5
0Correlation Response for 168 subcarriers in 6 MHz
Precise time samples
Am
plitu
des
(dB
)
8 µs 8 µs
Gerald Chouinard, CRCSlide 39
doc.: IEEE 802.22-11/0022r0
Submission
February 2011
Channel echo function for 84 subcarriersspread over 5.63 MHz (SNR = 6 dB)
1000 2000 3000 4000 5000 6000 7000 8000 9000 10000-0.4
-0.2
0
0.2
0.4
0.6
0.8
1Correlation Response for 84 subcarriers in 6 MHz
Precise time samples
Am
plitu
des
HR Chan. FunctionCCIR Abs samples
1000 2000 3000 4000 5000 6000 7000 8000 9000 10000-50
-45
-40
-35
-30
-25
-20
-15
-10
-5
0Correlation Response for 84 subcarriers in 6 MHz
Precise time samples
Am
plitu
des
(dB
)
8 µs 8 µs
Gerald Chouinard, CRCSlide 40
doc.: IEEE 802.22-11/0022r0
Submission
February 2011
Channel echo function for 84 subcarriersspread over 3.94 MHz (SNR = 6 dB)
8 µs 8 µs
1000 2000 3000 4000 5000 6000 7000 8000 9000 10000-50
-45
-40
-35
-30
-25
-20
-15
-10
-5
0Correlation Response for 84 subcarriers in 4.2 MHz
Precise time samples
Am
plitu
des
(dB
)
1000 2000 3000 4000 5000 6000 7000 8000 9000 10000-0.4
-0.2
0
0.2
0.4
0.6
0.8
1Correlation Response for 84 subcarriers in 4.2 MHz
Precise time samples
Am
plitu
des
HR Chan. FunctionCCIR Abs samples
Gerald Chouinard, CRCSlide 41
doc.: IEEE 802.22-11/0022r0
Submission
February 2011
Channel echo function for 84 subcarriers spreadover 2.8 MHz (SNR = 6 dB)
1000 2000 3000 4000 5000 6000 7000 8000 9000 10000-0.4
-0.2
0
0.2
0.4
0.6
0.8
1Correlation Response for 84 subcarriers in 3 MHz
Precise time samples
Am
plitu
des
HR Chan. FunctionCCIR Abs samples
1000 2000 3000 4000 5000 6000 7000 8000 9000 10000-50
-45
-40
-35
-30
-25
-20
-15
-10
-5
0Correlation Response for 84 subcarriers in 3 MHz
Precise time samples
Am
plitu
des
(dB
)
8 µs 8 µs
Gerald Chouinard, CRCSlide 42
doc.: IEEE 802.22-11/0022r0
Submission
February 2011
Channel echo function for 56 subcarriersspread over 3.75 MHz (SNR = 6 dB)
1000 2000 3000 4000 5000 6000 7000 8000 9000 10000-0.4
-0.2
0
0.2
0.4
0.6
0.8
1Correlation Response for 56 subcarriers in 4 MHz
Precise time samples
Am
plitu
des
HR Chan. FunctionCCIR Abs samples
1000 2000 3000 4000 5000 6000 7000 8000 9000 10000-50
-45
-40
-35
-30
-25
-20
-15
-10
-5
0Correlation Response for 56 subcarriers in 4 MHz
Precise time samples
Am
plitu
des
(dB
)
8 µs 8 µs
Gerald Chouinard, CRCSlide 43
doc.: IEEE 802.22-11/0022r0
Submission
February 2011
Channel echo function for 56 subcarriersspread over 2.63 MHz (SNR = 6 dB)
1000 2000 3000 4000 5000 6000 7000 8000 9000 10000-50
-45
-40
-35
-30
-25
-20
-15
-10
-5
0Correlation Response for 56 subcarriers in 2.8 MHz
Precise time samples
Am
plitu
des
(dB
)
8 µs 8 µs
1000 2000 3000 4000 5000 6000 7000 8000 9000 10000-0.4
-0.2
0
0.2
0.4
0.6
0.8
1Correlation Response for 56 subcarriers in 2.8 MHz
Precise time samples
Am
plitu
des
HR Chan. FunctionCCIR Abs samples
Gerald Chouinard, CRCSlide 44
doc.: IEEE 802.22-11/0022r0
Submission
February 2011
Channel echo function for 56 subcarriers spreadover 2.8 MHz (irregular pattern) (SNR = 6 dB)
1000 2000 3000 4000 5000 6000 7000 8000 9000 10000-0.4
-0.2
0
0.2
0.4
0.6
0.8
1Correlation Response for 56 subcarriers in 3 MHz
Precise time samples
Am
plitu
des
HR Chan. FunctionCCIR Abs samples
1000 2000 3000 4000 5000 6000 7000 8000 9000 10000-50
-45
-40
-35
-30
-25
-20
-15
-10
-5
0Correlation Response for 56 subcarriers in 3 MHz
Precise time samples
Am
plitu
des
(dB
)
8 µs 8 µs
Gerald Chouinard, CRCSlide 45
doc.: IEEE 802.22-11/0022r0
Submission
February 2011
Channel echo function for 56 subcarriersspread over 1.88 MHz (SNR = 6 dB)
1000 2000 3000 4000 5000 6000 7000 8000 9000 10000-0.4
-0.2
0
0.2
0.4
0.6
0.8
1Correlation Response for 56 subcarriers in 2 MHz
Precise time samples
Am
plitu
des
HR Chan. FunctionCCIR Abs samples
1000 2000 3000 4000 5000 6000 7000 8000 9000 10000-50
-45
-40
-35
-30
-25
-20
-15
-10
-5
0Correlation Response for 56 subcarriers in 2 MHz
Precise time samples
Am
plitu
des
(dB
)
8 µs 8 µs
Gerald Chouinard, CRCSlide 46
doc.: IEEE 802.22-11/0022r0
Submission
February 2011
Outline1. Extraction of the Complex Channel Impulse Response2. Generation of the Prototype Function
(i.e., High-resolution Complex Channel Impulse Response)
3. Channel deconvolution process4. Some subcarrier patterns for channel deconvolution
• Examples of perfect Channel Impulse Responses• Examples of Prototype Functions• Resulting Channel Echo Functions for various subcarrier patterns
5. Summary of the exercise6. Summary of the results7. Observations on the results8. Proposal
Gerald Chouinard, CRCSlide 47
doc.: IEEE 802.22-11/0022r0
Submission
February 2011
Summary of the exercise
• The channel deconvolution process described in previous contributions was used to establish the relative performance of various sets of static subcarriers used on the upstream for the channel profile model B.
• Beyond the localization performance of the various prototype functions resulting from different sets of static subcarriers, the resulting Channel Echo Functions were studied.
• The performance of the Channel Echo Functions were analyzed in terms of the width of the impulses representing the channel echoes and the residual ringing where peaks could be interpreted as false echoes.
• Minimum false echo rejection values were noted to establish the dynamic range of the process to identify real echoes before false echoes, resulting from unwanted ringing, start to appear.
• Simulations were first done without noise but were repeated with SNR= 6 dB (i.e., the minimum SNR for operation of the system at QPSK, rate: 1/2) to verify the effect of additive white Gaussian noise.
Gerald Chouinard, CRCSlide 48
doc.: IEEE 802.22-11/0022r0
Submission
February 2011
Outline1. Extraction of the Complex Channel Impulse Response2. Generation of the Prototype Function
(i.e., High-resolution Complex Channel Impulse Response)
3. Channel deconvolution process4. Some subcarrier patterns for channel deconvolution
• Examples of perfect Channel Impulse Responses• Examples of Prototype Functions• Resulting Channel Echo Functions for various subcarrier patterns
5. Summary of the exercise6. Summary of the results7. Observations on the results8. Proposal
Gerald Chouinard, CRCSlide 49
doc.: IEEE 802.22-11/0022r0
Submission
February 2011
Summary of the results of the 802.22 upstreamChannel Echo Response as a function of the
various static subcarrier sets
Number of sub-
channels
Number of sub-carriers
Subcarrier spacing
Bandwidth (MHz)
Max. echo delay range without aliasing (usec)
Main lobe width at -6 dB (micro-
samples)
Main lobe width at
-6 dB (ns)
Min. false echo rejection
(SNR=60 dB) (dB)
Min. false echo rejection
(SNR=6 dB) (dB)
60 1680 1 5.629 298.7 --- --- --- ---30 840 2 5.629 149.3 117.3 95.0 30.65 29.186 168 10 5.629 29.9 115.2 93.3 29.55 22.1 and 24.093 84 10 2.814 29.9 118.1 95.7 11 9.93 84 12 3.377 24.9 117.1 94.9 11.77 11.243 84 14 3.940 21.3 116.8 94.6 14.32 14.873 84 16 4.503 18.7 116.9 94.7 19.6 17.373 84 18 5.066 16.6 113.2 91.7 21.18 17.73 84 20 5.629 14.9 115.8 93.8 28.98 20.92 56 10 1.876 29.9 118.8 96.2 4.2 4.172 56 10 (irregular) 2.814 29.9 116.9 94.7 7.28 72 56 12 2.251 24.9 118.6 96.1 7.27 7.452 56 14 2.627 21.3 118.3 95.9 9.55 9.52 56 16 3.002 18.7 117.5 95.2 12.17 12.122 56 18 3.377 16.6 117.1 94.8 11.4 10.542 56 20 3.752 14.9 116.8 94.6 16.27 16.432 56 30 5.629 10.0 111.5 90.4 22.95 16.75
Gerald Chouinard, CRCSlide 50
doc.: IEEE 802.22-11/0022r0
Submission
February 2011
The iterative process to identify echoes• The exercise was repeated by using an iterative process
where each echo for which the amplitude and delay was identified was removed for the Channel Impulse Response to improve the detection performance of the process.
1. Largest echo, normalized to 1, is localized in the Channel Echo Function obtained from the correlation process (amplitude and delay are noted)
2. The values of the prototype function corresponding to this maximum echo are subtracted from the Channel Impulse Response
3. The resulting new Channel Impulse Response is correlated with the Prototype Function
4. The resulting new Channel Echo Function is scaled up to bring the second largest echo to an amplitude of 1 (scaling factor is recorded to note the amplitude difference)
5. Steps 1-4 are repeated until all echoes have been removed and what remains is noise.
6. The concatenated scaling factors represent the dynamic range of the iterative process.
Gerald Chouinard, CRCSlide 51
doc.: IEEE 802.22-11/0022r0
Submission
February 2011
Iterative echo detection process
0.2 0.4 0.6 0.8 1 1.2 1.4 1.6 1.8 2
x 104
-1
-0.8
-0.6
-0.4
-0.2
0
0.2
0.4
0.6
0.8
1Correlation Response for 84 subcarriers in 3.6 MHz
Precise time samples
Am
plitu
des
84 subcarriers regularly spaced every 12 subcarriers
over 3.38 MHz
Gerald Chouinard, CRCSlide 52
doc.: IEEE 802.22-11/0022r0
Submission
February 2011
Iterative echo detection process
0.2 0.4 0.6 0.8 1 1.2 1.4 1.6 1.8 2
x 104
-1
-0.8
-0.6
-0.4
-0.2
0
0.2
0.4
0.6
0.8
1Correlation Response for 84 subcarriers in 3.6 MHz
Precise time samples
Am
plitu
des
10 20 30 40 50 60 70 80 90 100 110 1200
0.1
0.2
0.3
0.4
0.5
0.6
0.7
0.8
0.9
1Subtraction of echo
Time samples
Rel
ativ
e am
plitu
de (d
B)
84 subcarriers regularly spaced every 12 subcarriers
over 3.38 MHz
Gerald Chouinard, CRCSlide 53
doc.: IEEE 802.22-11/0022r0
Submission
February 2011
Iterative echo detection process
0.2 0.4 0.6 0.8 1 1.2 1.4 1.6 1.8 2
x 104
-1
-0.8
-0.6
-0.4
-0.2
0
0.2
0.4
0.6
0.8
1Correlation Response for 84 subcarriers in 3.6 MHz
Precise time samples
Am
plitu
des
10 20 30 40 50 60 70 80 90 100 110 1200
0.1
0.2
0.3
0.4
0.5
0.6
0.7
0.8
0.9
1Subtraction of echo
Time samples
Rel
ativ
e am
plitu
de (d
B)
84 subcarriers regularly spaced every 12 subcarriers
over 3.38 MHz
Gerald Chouinard, CRCSlide 54
doc.: IEEE 802.22-11/0022r0
Submission
February 2011
Iterative echo detection process
0.2 0.4 0.6 0.8 1 1.2 1.4 1.6 1.8 2
x 104
-1
-0.8
-0.6
-0.4
-0.2
0
0.2
0.4
0.6
0.8
1Correlation Response for 84 subcarriers in 3.6 MHz
Precise time samples
Am
plitu
des
10 20 30 40 50 60 70 80 90 100 110 1200
0.1
0.2
0.3
0.4
0.5
0.6
0.7
0.8
0.9
1Subtraction of echo
Time samples
Rel
ativ
e am
plitu
de (d
B)
84 subcarriers regularly spaced every 12 subcarriers
over 3.38 MHz
Gerald Chouinard, CRCSlide 55
doc.: IEEE 802.22-11/0022r0
Submission
February 2011
Iterative echo detection process
0.2 0.4 0.6 0.8 1 1.2 1.4 1.6 1.8 2
x 104
-1
-0.8
-0.6
-0.4
-0.2
0
0.2
0.4
0.6
0.8
1Correlation Response for 84 subcarriers in 3.6 MHz
Precise time samples
Am
plitu
des
10 20 30 40 50 60 70 80 90 100 110 1200
0.1
0.2
0.3
0.4
0.5
0.6
0.7
0.8
0.9
1Subtraction of echo
Time samples
Rel
ativ
e am
plitu
de (d
B)
84 subcarriers regularly spaced every 12 subcarriers
over 3.38 MHz
Gerald Chouinard, CRCSlide 56
doc.: IEEE 802.22-11/0022r0
Submission
February 2011
Iterative echo detection process
0.2 0.4 0.6 0.8 1 1.2 1.4 1.6 1.8 2
x 104
-1
-0.8
-0.6
-0.4
-0.2
0
0.2
0.4
0.6
0.8
1Correlation Response for 84 subcarriers in 3.6 MHz
Precise time samples
Am
plitu
des
10 20 30 40 50 60 70 80 90 100 110 1200
0.1
0.2
0.3
0.4
0.5
0.6
0.7
0.8
0.9
1Subtraction of echo
Time samples
Rel
ativ
e am
plitu
de (d
B)
84 subcarriers regularly spaced every 12 subcarriers
over 3.38 MHz
Gerald Chouinard, CRCSlide 57
doc.: IEEE 802.22-11/0022r0
Submission
February 2011
Iterative echo detection process
0.2 0.4 0.6 0.8 1 1.2 1.4 1.6 1.8 2
x 104
-1
-0.8
-0.6
-0.4
-0.2
0
0.2
0.4
0.6
0.8
1Correlation Response for 84 subcarriers in 3.6 MHz
Precise time samples
Am
plitu
des
10 20 30 40 50 60 70 80 90 100 110 1200
0.1
0.2
0.3
0.4
0.5
0.6
0.7
0.8
0.9
1Subtraction of echo
Time samples
Rel
ativ
e am
plitu
de (d
B)
84 subcarriers regularly spaced every 12 subcarriers
over 3.38 MHz Left over of earlier detected
echoes
Gerald Chouinard, CRCSlide 58
doc.: IEEE 802.22-11/0022r0
Submission
February 2011
Iterative echo detection process
0.2 0.4 0.6 0.8 1 1.2 1.4 1.6 1.8 2
x 104
-1
-0.8
-0.6
-0.4
-0.2
0
0.2
0.4
0.6
0.8
1Correlation Response for 84 subcarriers in 3.6 MHz
Precise time samples
Am
plitu
des
10 20 30 40 50 60 70 80 90 100 110 1200
0.1
0.2
0.3
0.4
0.5
0.6
0.7
0.8
0.9
1Subtraction of echo
Time samples
Rel
ativ
e am
plitu
de (d
B)
84 subcarriers regularly spaced every 12 subcarriers
over 3.38 MHz Left over of the earlier detected
echoes
Gerald Chouinard, CRCSlide 59
doc.: IEEE 802.22-11/0022r0
Submission
February 2011
Iterative echo detection process
0.2 0.4 0.6 0.8 1 1.2 1.4 1.6 1.8 2
x 104
-1
-0.8
-0.6
-0.4
-0.2
0
0.2
0.4
0.6
0.8
1Correlation Response for 84 subcarriers in 3.6 MHz
Precise time samples
Am
plitu
des
10 20 30 40 50 60 70 80 90 100 110 1200
0.1
0.2
0.3
0.4
0.5
0.6
0.7
0.8
0.9
1Subtraction of echo
Time samples
Rel
ativ
e am
plitu
de (d
B)
84 subcarriers regularly spaced every 12 subcarriers
over 3.38 MHz Left over of the earlier detected
echoes
Gerald Chouinard, CRCSlide 60
doc.: IEEE 802.22-11/0022r0
Submission
February 2011
Iterative echo detection process
0.2 0.4 0.6 0.8 1 1.2 1.4 1.6 1.8 2
x 104
-1
-0.8
-0.6
-0.4
-0.2
0
0.2
0.4
0.6
0.8
1Correlation Response for 84 subcarriers in 3.6 MHz
Precise time samples
Am
plitu
des
10 20 30 40 50 60 70 80 90 100 110 1200
0.1
0.2
0.3
0.4
0.5
0.6
0.7
0.8
0.9
1Subtraction of echo
Time samples
Rel
ativ
e am
plitu
de (d
B)
84 subcarriers regularly spaced every 12 subcarriers
over 3.38 MHz
Gerald Chouinard, CRCSlide 61
doc.: IEEE 802.22-11/0022r0
Submission
February 2011
Iterative echo detection process
0.2 0.4 0.6 0.8 1 1.2 1.4 1.6 1.8 2
x 104
-1
-0.8
-0.6
-0.4
-0.2
0
0.2
0.4
0.6
0.8
1Correlation Response for 56 subcarriers in 3 MHz
Precise time samples
Am
plitu
des
56 subcarriers irregularly spaced every 10 subcarriers
over 2.8 MHz
Gerald Chouinard, CRCSlide 62
doc.: IEEE 802.22-11/0022r0
Submission
February 2011
Iterative echo detection process
0.2 0.4 0.6 0.8 1 1.2 1.4 1.6 1.8 2
x 104
-1
-0.8
-0.6
-0.4
-0.2
0
0.2
0.4
0.6
0.8
1Correlation Response for 56 subcarriers in 3 MHz
Precise time samples
Am
plitu
des
10 20 30 40 50 60 70 80 90 100 110 1200
0.1
0.2
0.3
0.4
0.5
0.6
0.7
0.8
0.9
1Subtraction of echo
Time samples
Rel
ativ
e am
plitu
de (d
B)
56 subcarriers irregularly spaced every 10 subcarriers
over 2.8 MHz
Gerald Chouinard, CRCSlide 63
doc.: IEEE 802.22-11/0022r0
Submission
February 2011
Iterative echo detection process
0.2 0.4 0.6 0.8 1 1.2 1.4 1.6 1.8 2
x 104
-1
-0.8
-0.6
-0.4
-0.2
0
0.2
0.4
0.6
0.8
1Correlation Response for 56 subcarriers in 3 MHz
Precise time samples
Am
plitu
des
10 20 30 40 50 60 70 80 90 100 110 1200
0.1
0.2
0.3
0.4
0.5
0.6
0.7
0.8
0.9
1Subtraction of echo
Time samples
Rel
ativ
e am
plitu
de (d
B)
56 subcarriers irregularly spaced every 10 subcarriers
over 2.8 MHz
Gerald Chouinard, CRCSlide 64
doc.: IEEE 802.22-11/0022r0
Submission
February 2011
Iterative echo detection process
0.2 0.4 0.6 0.8 1 1.2 1.4 1.6 1.8 2
x 104
-1
-0.8
-0.6
-0.4
-0.2
0
0.2
0.4
0.6
0.8
1Correlation Response for 56 subcarriers in 3 MHz
Precise time samples
Am
plitu
des
10 20 30 40 50 60 70 80 90 100 110 1200
0.1
0.2
0.3
0.4
0.5
0.6
0.7
0.8
0.9
1Subtraction of echo
Time samples
Rel
ativ
e am
plitu
de (d
B)
56 subcarriers irregularly spaced every 10 subcarriers
over 2.8 MHz
Gerald Chouinard, CRCSlide 65
doc.: IEEE 802.22-11/0022r0
Submission
February 2011
Iterative echo detection process
0.2 0.4 0.6 0.8 1 1.2 1.4 1.6 1.8 2
x 104
-1
-0.8
-0.6
-0.4
-0.2
0
0.2
0.4
0.6
0.8
1Correlation Response for 56 subcarriers in 3 MHz
Precise time samples
Am
plitu
des
10 20 30 40 50 60 70 80 90 100 110 1200
0.1
0.2
0.3
0.4
0.5
0.6
0.7
0.8
0.9
1Subtraction of echo
Time samples
Rel
ativ
e am
plitu
de (d
B)
56 subcarriers irregularly spaced every 10 subcarriers
over 2.8 MHz
Gerald Chouinard, CRCSlide 66
doc.: IEEE 802.22-11/0022r0
Submission
February 2011
Number of sub-
channels
Number of sub-carriers
Subcarrier spacing
Bandwidth (MHz)
Max. echo delay range without aliasing (usec)
Main lobe width at -6 dB (micro-
samples)
Main lobe width at
-6 dB (ns)
Min. false echo rejection
(SNR=60 dB) (dB)
Min. false echo rejection
(SNR=6 dB) (dB)
Min. false echo rejection
(SNR=60 dB) (dB)
Min. false echo rejection
(SNR=6 dB) (dB)
60 1680 1 5.629 298.7 --- --- --- ---30 840 2 5.629 149.3 117.3 95.0 30.65 29.18 41.07 26.946 168 10 5.629 29.9 115.2 93.3 29.55 22.1 and 24.09 33.21 22.383 84 10 2.814 29.9 118.1 95.7 11 9.9 27.24 16.823 84 12 3.377 24.9 117.1 94.9 11.77 11.24 34.30 18.543 84 14 3.940 21.3 116.8 94.6 14.32 14.87 33.92 20.623 84 16 4.503 18.7 116.9 94.7 19.6 17.37 39.59 19.183 84 18 5.066 16.6 113.2 91.7 21.18 17.7 38.24 20.223 84 20 5.629 14.9 115.8 93.8 28.98 20.92 56 10 1.876 29.9 118.8 96.2 4.2 4.17 28.02 17.012 56 10 (irregular) 2.814 29.9 116.9 94.7 7.28 7 10.12 10.012 56 12 2.251 24.9 118.6 96.1 7.27 7.45 27.11 17.222 56 14 2.627 21.3 118.3 95.9 9.55 9.5 35.01 17.092 56 16 3.002 18.7 117.5 95.2 12.17 12.12 26.23 16.222 56 18 3.377 16.6 117.1 94.8 11.4 10.54 25.08 17.842 56 20 3.752 14.9 116.8 94.6 16.27 16.432 56 30 5.629 10.0 111.5 90.4 22.95 16.75
Iterative process
Summary of the results of the 802.22 upstreamChannel Echo Response as a function of the
various static subcarrier sets
Gerald Chouinard, CRCSlide 67
doc.: IEEE 802.22-11/0022r0
Submission
February 2011
Number of sub-
channels
Number of sub-carriers
Subcarrier spacing
Bandwidth (MHz)
Max. echo delay range without aliasing (usec)
Min. false echo rejection
(SNR=60 dB) (dB)
Min. false echo rejection
(SNR=6 dB) (dB)
Min. false echo rejection
(SNR=60 dB) (dB)
Min. false echo rejection
(SNR=6 dB) (dB)
% Bandwidth
% Echo range
60 1680 1 5.629 298.7 --- ---30 840 2 5.629 149.3 30.65 29.18 41.07 26.946 168 10 5.629 29.9 29.55 22.1 and 24.09 33.21 22.383 84 10 2.814 29.9 11 9.9 27.24 16.82 50% 100%3 84 12 3.377 24.9 11.77 11.24 34.30 18.54 60% 83%3 84 14 3.940 21.3 14.32 14.87 33.92 20.62 70% 71%3 84 16 4.503 18.7 19.6 17.37 39.59 19.18 80% 63%3 84 18 5.066 16.6 21.18 17.7 38.24 20.22 90% 56%3 84 20 5.629 14.9 28.98 20.92 56 10 1.876 29.9 4.2 4.17 28.02 17.01 33% 100%2 56 10 (irregular) 2.814 29.9 7.28 7 10.12 10.01 50% 100%2 56 12 2.251 24.9 7.27 7.45 27.11 17.22 40% 83%2 56 14 2.627 21.3 9.55 9.5 35.01 17.09 47% 71%2 56 16 3.002 18.7 12.17 12.12 26.23 16.22 53% 63%2 56 18 3.377 16.6 11.4 10.54 25.08 17.84 60% 56%2 56 20 3.752 14.9 16.27 16.432 56 30 5.629 10.0 22.95 16.75
Iterative process
Summary of the results of the 802.22 upstreamChannel Echo Response as a function of the
various static subcarrier sets
Gerald Chouinard, CRCSlide 68
doc.: IEEE 802.22-11/0022r0
Submission
February 2011
Outline1. Extraction of the Complex Channel Impulse Response2. Generation of the Prototype Function
(i.e., High-resolution Complex Channel Impulse Response)
3. Channel deconvolution process4. Some subcarrier patterns for channel deconvolution
• Examples of perfect Channel Impulse Responses• Examples of Prototype Functions• Resulting Channel Echo Functions for various subcarrier patterns
5. Summary of the exercise6. Summary of the results7. Observations on the results8. Proposal
Gerald Chouinard, CRCSlide 69
doc.: IEEE 802.22-11/0022r0
Submission
February 2011
Observations on the results
• Effective bandwidth occupied by static subcarriers– The effective bandwidth over which the static subcarriers are spread has only a small
influence on the width of the impulses in the Channel Echo Function. In other words, the accuracy achievable on the precise time position of the echoes reduces only slightly with a reduction of the effective bandwidth, i.e., the accuracy of the precise echo delays is preserved even in the case of narrower bandwidths.
– However, it should be remembered that larger effective bandwidths will reduce the effect of echo smearing due to the Nyquist limitation related to the capability to differentiate closely spaced echoes for a given signal bandwidth, i.e., echo discrimination and exact delay will be better preserved with a larger signal bandwidth.
• Maximum echo delay range that can be characterized without aliasing– The maximum echo delay range that can be covered without aliasing is linearly related
to the static subcarrier spacing: larger the spacing is, shorter the maximum delay range is.
– The delay ranges covered by the 802.22 system for cyclic prefix 1/4, 1/8, 1/16 and 1/32 are: 75, 37.5, 18.75 and 9.4 µs respectively. The delay range covered by the downstream preamble is 149.3 µs. This is to be compared to the delay ranges resulting from the various upstream training options which vary from 10 µs to 30 µs.
Gerald Chouinard, CRCSlide 70
doc.: IEEE 802.22-11/0022r0
Submission
February 2011
Observations on the results (Cont’d)
• Tradeoff between occupied bandwidth and available delay range– There is a tradeoff, for a given number of sub-channels allocated to static
subcarriers, between the effective bandwidth over which the static subcarriers are spread and the maximum echo delay range that can be characterized without aliasing
• Regular versus irregular subcarrier spacing– Regular subcarrier spacing give better and more predictable sidelobe levels as
compared to irregularly sub-sampled subcarriers in the frequency domain (see the second row of the 2 sub-channel group on slide 49).
• False echo rejection performance– The downstream LTS preamble occupying 30 sub-channels offers some 30 dB false
echo rejection (41 dB with the iterative process)– If 6 sub-channels were used in the upstream for static subcarriers, the false echo
rejection would still be around 30 dB but the rejection would decrease to 22-24 dB in presence of large amount of noise (SNR= 6 dB) (33 dB and 22 dB respectively with the iterative process)
Gerald Chouinard, CRCSlide 71
doc.: IEEE 802.22-11/0022r0
Submission
February 2011
Observations on the results (Cont’d)
• False echo rejection performance (cont’d)– The false echo rejection reduces from 29 dB to 11 dB for 3 sub-channels with the
decrease in subcarrier spacing. However, the highest values are achieved at the cost of a reduced maximum echo delay range. Lower rejection values are found when noise is present.
– The false echo rejection reduces from 16.3 dB to 4.2 dB for 2 sub-channels with the decrease in subcarrier spacing. Again, the highest values are achieved at the cost of a reduced maximum echo delay range (down to 15 delay range, a 10 µs delay range example was included for convenience but is clearly insufficient). Slightly lower rejection values are found when noise is present.
– However, when an iterative process is used to identify echoes (once an echo is identified, it is removed from the Channel Impulse Function to ease the identification of the next one), much better false echo performance can be achieved (between 25 and 40 dB) when using regularly spaced subcarriers.
– Such performance is limited at low SNR (SNR= 6 dB is used as the threshold fpr proper operation of the 802.22 WRAN system). Performance in the range of 16 to 20 dB can be achieved with the iterative process.
– Irregularly spaced ranging subcarriers has a limited performance even in the case of the iterative echo removal process (limited to 10 dB range independent of noise).
Gerald Chouinard, CRCSlide 72
doc.: IEEE 802.22-11/0022r0
Submission
February 2011
Outline1. Extraction of the Complex Channel Impulse Response2. Generation of the Prototype Function
(i.e., High-resolution Complex Channel Impulse Response)
3. Channel deconvolution process4. Some subcarrier patterns for channel deconvolution
• Examples of perfect Channel Impulse Responses• Examples of Prototype Functions• Resulting Channel Echo Functions for various subcarrier patterns
5. Summary of the exercise6. Summary of the results7. Observations on the results8. Proposal
Gerald Chouinard, CRCSlide 73
doc.: IEEE 802.22-11/0022r0
Submission
February 2011
Proposal• It is proposed to use 3 sub-channels for the upstream CDMA static
carriers, with the following parameters:– Total number of subcarriers = 84– Subcarrier spacing = 16
• The performance of this upstream static subcarrier arrangement is:– Total effective bandwidth = 4.5 MHz (i.e., 80% of total bandwidth giving sufficient
protection against echo smearing)– Maximum echo delay range (before aliasing starts to appear) = 18.7 µs
(i.e., 63% of the 30 µsec echo delay range)
– Minimum false echo rejection = 19.6 dB (17.4 dB at SNR = 6 dB) 39.6 dB (19.2 dB at SNR = 6 dB) with
iterations• It is proposed to make the necessary modifications to section 8.6 of the
P802.22 D1.0 to reserve 3 sub-channels for the static subcarriers on the upstream and indicate the exact location of these static subcarriers in Table 201.
Gerald Chouinard, CRCSlide 74
doc.: IEEE 802.22-11/0022r0
Submission
February 2011
References1. IEEE P802.22™/ DRAFTv7.0 Draft Standard for Wireless Regional Area Networks Part 22:
Cognitive Wireless RAN Medium Access Control (MAC) and Physical Layer (PHY) specifications: Policies and procedures for operation in the TV Bands, December 2009
2. 22-06-0206-00-0000-ranging-with-ofdm-systems.ppt3. 22-10-0055-0000-Multicarrier-ranging.ppt4. 22-10-0054-02-0000_OFDM-based Terrestrial Geolocation.ppt5. 22-10-0178-01-0000 Updated set of CDMA ranging tones.doc