1. Adjacent Channel Interference and Filtering for 56-carrier Signals
January 2005
doc.: IEEE /1579r0
January 2005
Adjacent Channel Interference and Filtering for 56-carrier Signals
Date:
Authors:
Notice: This document has been prepared to assist IEEE 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
Patent Policy and Procedures: The contributor is familiar with the IEEE 802 Patent Policy and Procedures < ieee802.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 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 Working Group. If you have questions, contact the IEEE Patent Committee Administrator at
Dave Hedberg, Conexant Systems
Dave Hedberg, Conexant Systems

2. January 2005
doc.: IEEE /1579r0
January 2005
Abstract
Proposed TGn signal tone structures utilize added sub-carriers to increase data rates. The consequent larger fractional signal bandwidth and smaller separation between neighboring channel signals has raised concerns about adjacent channel interference and filtering difficulty. This presentation shows that with appropriate Rx filter choices, the adjacent channel rejection performance and Rx filter complexity do not change substantially for 20 MHz 56-carrier WWiSE signals vs. the standard 52-carrier signals. Examples and simulations are shown for both cases to show comparable ACI performance with minimal filter change and minimal complexity impact. Likewise, the Tx adjacent channel leakage performance due to PA distortion for 56-carriers is for all practical purposes the same as for current standard 52-carrier signals. In typical implementations the DSP filter lengths must be increased for 56-carriers, but the impact on performance and complexity is minimal. It is also shown that added dispersion due to the longer Tx/Rx filter impulse responses for 56-carriers will not significantly change PER performance with TGn channel models.
Dave Hedberg, Conexant Systems
Dave Hedberg, Conexant Systems

3. Outline Basic Model and Observations ACI, AACI Mechanisms
January 2005
Outline
Basic Model and Observations
ACI, AACI Mechanisms
Receiver Aliasing of Minimally Distorted ACI Signals
Filter Examples and ACI Comparisons: 56 vs. 52 carriers
Filter Dispersion
Interference with Maximally Distorted ACI Signals
Conclusions
Dave Hedberg, Conexant Systems

4. 20 MHz Channel Spectra - Aliasing
January 2005
20 MHz Channel Spectra - Aliasing
From a
Specification
Dave Hedberg, Conexant Systems

5. January 2005
Observations
Let fractional bandwidth, a, be defined as the total sub-carrier
bandwidth span divided by the channel Nyquist bandwidth
For the model shown on previous slide, define the normalized
filter transition width parameter as
b = 2 ( 1 - a)
Rx filter steepness (and complexity) are inversely related to b
Signal Model a
a3dB b
52- carrier .11a/g
0.8125
0.828
0.375
56-carrier WWiSE
0.8750
0.891
0.250
114-carrier TGnSync
0.914
0.1875
112-carrier WWiSE
Dave Hedberg, Conexant Systems

6. ACI Mechanisms and Metrics
January 2005
ACI Mechanisms and Metrics
Aliasing in receiver of incompletely filtered out-of-channel signals (Adjacent Channel Rejection, ACR)
Average total adjacent channel signal power aliased into "active channel" FFT bins
Minimum sub-carrier alias rejection (typically at outer sub-carrier)
Bleeding of an adjacent channel Tx distortion shoulder into "active channel" FFT bins (Adjacent Channel Leakage, ACL)
Average total adjacent channel signal power overlapping into "active
channel" FFT bins
ACI Power Measure:
Total ACI power is normalized to total in-band power of the adjacent signal seen at the receiver input
Adjacent channel interferer is assumed a signal of the same type
Dave Hedberg, Conexant Systems

7. January 2005
Adjacent Channel Rejection with Minimally Distorted ACI Signals - Filter Examples
Dave Hedberg, Conexant Systems

8. Filter Examples - 2x OSR Rx Path
January 2005
Filter Examples - 2x OSR Rx Path
Analog anti-aliasing filter should be at least 5 – poles to meet AACR performance, e.g. with conventional analog bi-quad filters
Assume analog AA filter corner frequency calibration accuracy of +/-5%
Over-sampling allows transition edge characteristic to be determined
by a precise digital filter – typical in current radio technology
Half-band digital decimation filters are often used because of efficiency
Dave Hedberg, Conexant Systems

9. Example 5-pole Cheby AA Filter - Limits
January 2005
Example 5-pole Cheby AA Filter - Limits
Dave Hedberg, Conexant Systems

10. Composite Filter Responses and Rejection for .11a/g 52-carrier Example
January 2005
Composite Filter Responses and Rejection for .11a/g 52-carrier Example
Dave Hedberg, Conexant Systems

11. Composite Filter Responses and Rejection for WWiSE 56-carrier Example
January 2005
Composite Filter Responses and Rejection for WWiSE 56-carrier Example
Dave Hedberg, Conexant Systems

12. January 2005
Receive ACI Spectra with Minimally Distorted 56 and 52 carrier Neighbors
ACI Signal Located in Adjacent Channel
ACI Signal Shifted to Receive Band
Max outer carrier aliasing (at index 5 for WWiSE 56-carrier signal, index 7 for 52-carrier signal) with example filters
Dave Hedberg, Conexant Systems

13. January 2005
Receive ACI Spectrum with Minimally Distorted 56-carrier Neighbor + Freq Offset
ACI Signal Located in Adjacent Channel
Max outer carrier aliasing at index 5
Dave Hedberg, Conexant Systems

14. Group Delay Comparison
January 2005
Group Delay Comparison
Group delay differences for 56-carrier vs. 52-carrier filter examples
Dave Hedberg, Conexant Systems

15. January 2005
Back-up Slide
Dave Hedberg, Conexant Systems

16. Dispersion Due to Filtering vs. Channel Impulse Responses
January 2005
Dispersion Due to Filtering vs. Channel Impulse Responses
Total time dispersion of the signal at the Rx FFT detector should be comfortably less than the guard interval of 0.8 us.
The following simulation result measures the energy-time dispersion for a 56-carrier symbol spectrum in a flat channel vs. typical instances of TGn Ch D, NLOS
A receiver filter matched to the IFFT "impulse" symbol is used to detect the impulse response dispersion of everything present in the "channel", including the Tx and Rx filters
The resulting matched filter output is observed for cases of no Rx filter vs. the example Rx filters; and with and without channel multipath dispersion between the Tx and Rx
Dave Hedberg, Conexant Systems

17. Detected Rx Symbol Energy Dispersion
January 2005
Detected Rx Symbol Energy Dispersion No
Rx Filter
Flat Ch
Example
56-carrier
Rx Filter
5 Ch. D's No
Rx Filter
5 Ch. D's
56-carrier
Rx Filter
800 ns
Dave Hedberg, Conexant Systems

18. Filter Differences - 56 vs. 52 - carriers
January 2005
Filter Differences - 56 vs carriers
Analog anti-aliasing filter requirements do not change for 56-carrier signals – since they are driven primarily by the AACR requirements
e.g., the same 5-pole filter with +/-5% tolerance is adequate for both
WWiSE 56-carrier signals require slightly steeper digital filter for
ACR performance equivalent to that with 52-carrier signals
27-tap vs. 19-tap equivalent HBF length (2 physical tap difference)
Complexity difference is miniscule
Rx filter impact to channel dispersion is small in both cases (with small group delay variation contributed by the analog AA filter)
e.g. TGn channel D multipath dispersion is much larger than filter dispersion
Dave Hedberg, Conexant Systems

19. Adjacent Channel Leakage from Maximally Distorted ACI Signals
January 2005
Adjacent Channel Leakage from Maximally Distorted ACI Signals
Dave Hedberg, Conexant Systems

20. Maximally Distorted Tx Spectrum
January 2005
Maximally Distorted Tx Spectrum
56-carrier WWiSE Signal with 4 dB PA OBO
Dave Hedberg, Conexant Systems

21. Receive ACI Spectrum with Maximally Distorted 56-carrier Neighbors
January 2005
Receive ACI Spectrum with Maximally Distorted 56-carrier Neighbors
56-carrier WWiSE Signal – ACI shifted into band
4 dB PA OBO
ACI Signal Located in Adjacent Channel
Max outer carrier aliasing (at index 5) due to Rx filtering
(56-carrier filter example)
Dave Hedberg, Conexant Systems

22. Receive ACI Spectrum with Maximally Distorted 52-carrier Neighbors
January 2005
Receive ACI Spectrum with Maximally Distorted 52-carrier Neighbors
52-carrier Signal – ACI shifted into band
3.5 dB PA OBO
ACI Signal Located in Adjacent Channel
Max outer carrier aliasing (at index 7) due to Rx filtering
(52-carrier filter example)
Dave Hedberg, Conexant Systems

23. Comparisons of ACL for 56 vs. 52 carriers
January 2005
Comparisons of ACL for 56 vs. 52 carriers
ACL expressed in dB of leakage into passband of adjacent signal band
Rapp p=3.0 PA model
Distortion and leakage change by ~ 3 dB per 1 dB of PA OBO change
For all practical purposes, the Tx adjacent channel leakage performance
for 56-carriers is the same as for 52-carriers
Dave Hedberg, Conexant Systems

24. Simulation Block Diagram
January 2005
Simulation Block Diagram
Dave Hedberg, Conexant Systems

25. January 2005
Overall Conclusions
ACI performance and required Tx and Rx filter complexity are
not significantly different for 56-carriers vs. 52-carriers
With the "conservative" filter examples shown, PER
simulations show several dB of ACR and AACR margin for
20 MHz WWiSE MIMO modes relative to a/g 54 Mbps
ACR requirements
The added dispersion due to the required narrower filter
transition band for 56-carriers does not significantly impact
PER performance with TGn channels
Dave Hedberg, Conexant Systems

26. January 2005
References
IEEE / n, “WWiSE Proposal: High throughput extension to the Standard,” C. Hansen, B. Edwards et al.
IEEE / n, “WWiSE IEEE n Proposal,” S. Coffey et al., Nov. 16, 2004
Dave Hedberg, Conexant Systems