Doc.: IEEE 802.22-07/0357r0 Submission July 2007 Jay Unnikrishnan, QualcommSlide 1 Simulation of Eigenvalue based sensing of wireless mics IEEE P802.22.

July 2007

Jay Unnikrishnan, Qualcomm

Slide 1

doc.: IEEE 802.22-07/0357r0

Submission

Simulation of Eigenvalue based sensing of wireless mics

IEEE P802.22 Wireless RANs Date: 2007-07-16

Name Company Address Phone email Jayakrishnan Unnikrishnan

Qualcomm 5775 Morehouse Dr San Diego CA 92121

(858) 651-8327 [email protected]

Steve Shellhammer Qualcomm 5775 Morehouse Dr San Diego CA 92121

(858) 658-1874 [email protected]

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 802.22.

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 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 802.22 Working Group. If you have questions, contact the IEEE Patent Committee Administrator at [email protected].>

July 2007

Jay Unnikrishnan, Qualcomm

Slide 2

doc.: IEEE 802.22-07/0357r0

Submission

Abstract

• We simulated the Eigenvalue-based technique for sensing wireless microphone from document [1]

• Simulations were carried out for twelve different operating conditions of the transmitted wireless microphone signal as specified by Chris Clanton– A new model for the audio signal using colored noise was

introduced

July 2007

Jay Unnikrishnan, Qualcomm

Slide 3

doc.: IEEE 802.22-07/0357r0

Submission

Wireless microphone signals

• Most wireless microphones use analog FM modulation

• Signal is located in a 200 kHz sub-band within the 6 MHz TV channel

• Since the signal occupies only a narrow band within the TV channel, detection techniques that based on correlation in the signal – or equivalently spectral flatness – are expected to work well

July 2007

Jay Unnikrishnan, Qualcomm

Slide 4

doc.: IEEE 802.22-07/0357r0

Submission

Test signals

• Half of the test signal cases were generated following instructions in [2]

• Three different values of FM frequency deviation were used– 5 kHz (silent)

– 15 kHz (soft speaker)

– 32.6 kHz (loud speaker)

• Test audio signals in [2] are represented by tones

• Test signals with more accurate representations of audio signals are described later

July 2007

Jay Unnikrishnan, Qualcomm

Slide 5

doc.: IEEE 802.22-07/0357r0

Submission

Simulation setup

• Considered two channels– No fading

– A single tap with Rayleigh fading (although we are sampling at a high rate (~21 MHz), the signal of interest is narrowband (~ 100 kHz) and hence a single tap channel is sufficient)

• A 6 MHz pass-band filter is used at the receiver

• Following [1], processing is done at the same IF frequency as was used by DTV signals (5.381119 MHz) and same sampling rate as well (21.524476 MHz)

July 2007

Jay Unnikrishnan, Qualcomm

Slide 6

doc.: IEEE 802.22-07/0357r0

Submission

Frequency response of the pass-band filter

July 2007

Jay Unnikrishnan, Qualcomm

Slide 7

doc.: IEEE 802.22-07/0357r0

Submission

Spectrum Plot for case of Loud Speaker

July 2007

Jay Unnikrishnan, Qualcomm

Slide 8

doc.: IEEE 802.22-07/0357r0

Submission

Spectrum Plot for case of Loud Speaker

July 2007

Jay Unnikrishnan, Qualcomm

Slide 9

doc.: IEEE 802.22-07/0357r0

Submission

Maximum-minimum Eigenvalue-based sensing

• The algorithm suggested in [1] was implemented

• Sample autocorrelation matrix is calculated and transformed by multiplying with compensation matrices to whiten the spectrum of band-pass filtered noise

• Smoothing factor (size of the autocorrelation matrix) was chosen as L = 10

• Ratio of maximum Eigenvalue to minimum Eigenvalue of the compensated sample autocorrelation matrix is the test statistic

• Sensing time of 9.3 ms was used

• Threshold is set to achieve a false alarm rate of 10%

July 2007

Jay Unnikrishnan, Qualcomm

Slide 10

doc.: IEEE 802.22-07/0357r0

Submission

Flowchart of algorithm (taken from [1])

Transform the sample covariance matrix

Decision: if the maximum eign >r*minimum eign,signal exists;Otherwise, signal does not exist.

Choose a smoothing factor and the threshold r

Compute the maximum eigenvalue and minimum eigenvalue of the covariance matrix

Sample and filter the signals

Compute the sample covariance matrix

July 2007

Jay Unnikrishnan, Qualcomm

Slide 11

doc.: IEEE 802.22-07/0357r0

Submission

Simulations

• SNR values were varied and the probability of mis-detection was plotted against SNR

• SNR value required for PMD = 10% is the performance criterion of interest

• SNR is calculated at the output of the band-pass filter over the entire 6 MHz range of the TV channel since the exact location of the wireless microphone signal within the band is unknown a priori

July 2007

Jay Unnikrishnan, Qualcomm

Slide 12

doc.: IEEE 802.22-07/0357r0

Submission

Simulation Results – With and w/o Fading

July 2007

Jay Unnikrishnan, Qualcomm

Slide 13

doc.: IEEE 802.22-07/0357r0

Submission

Adjacent channel rejection

• Tested the effect of a wireless microphone signal in the adjacent channel

• Signal to noise (SNR) is measured as the ratio of the following– Signal power of the adjacent channel signal

– Noise in a 6 MHz bandwidth

July 2007

Jay Unnikrishnan, Qualcomm

Slide 14

doc.: IEEE 802.22-07/0357r0

Submission

Adjacent channel simulation results

July 2007

Jay Unnikrishnan, Qualcomm

Slide 15

doc.: IEEE 802.22-07/0357r0

Submission

New Audio Signal Model

• In addition to the tones suggested in [2] for representing the audio signal, more accurate models of voice signals were also simulated using the circuit from ETSI document no. ETSI EN 300 422-1 (V1.2.2)

• Audio signal is simulated using colored noise generated by passing white noise through the ETSI circuit

• Audio signal is then passed through a pre-emphasis filter (with time constant 400 us) prior to modulation

July 2007

Jay Unnikrishnan, Qualcomm

Slide 16

doc.: IEEE 802.22-07/0357r0

Submission

Filter circuit used to simulate audio signal

July 2007

Jay Unnikrishnan, Qualcomm

Slide 17

doc.: IEEE 802.22-07/0357r0

Submission

ETSI filter response and audio spectrum

July 2007

Jay Unnikrishnan, Qualcomm

Slide 18

doc.: IEEE 802.22-07/0357r0

Submission

Simulation Results – With and w/o Fading

July 2007

Jay Unnikrishnan, Qualcomm

Slide 19

doc.: IEEE 802.22-07/0357r0

Submission

Conclusions

• Performance of algorithm is not significantly affected by the FM deviation, the audio signal source or the level of pre-emphasis

• The band-pass filter is effective at suppressing adjacent channel signals

• Required SNR without fading is approximately -20 dB– Same as in [1]

• Required SNR with fading is approximately -12 dB– Different than in [1] (results from [1] approx -20 dB)

July 2007

Jay Unnikrishnan, Qualcomm

Slide 20

doc.: IEEE 802.22-07/0357r0

Submission

References

1. Yonghong Zeng and Ying-Chang Liang, “Performance of Eigenvalue based sensing algorithms for detection of DTV and wireless microphone signals,” 802.22-06/186r0, September 2006

2. Chris Clanton, Mark Kenkel and Yang Tang, “Wireless Microphone Signal Simulation Method”, 802.22-07/124r0, March 2007