March 2007

Monisha Ghosh, Philips

Slide 1

doc.: IEEE 802.22-07/0125-00-0000

Submission

DTV Signal Sensing Using Pilot Detection

IEEE P802.22 Wireless RANs Date: 2007-03-11

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 patcom@iee.org.>

Name Company Address Phone email Monisha Ghosh Philips USA 914-945-6415 Monisha.Ghosh@philips.com

Vasanth Gaddam Philips USA 914-945-6424 Vasanth.gaddam@philips.com

Gene Turkenich Philips USA 914-945-6370 Gene.turkenich@philips.com

March 2007

Monisha Ghosh, Philips

Slide 2

doc.: IEEE 802.22-07/0125-00-0000

Submission

Sensing the DTV pilot

• The DTV pilot is an unique feature that can be used to sense the presence of the DTV signal.

• Standard approach:– Narrow-band filter (~ 10 kHz) centered around nominal DTV pilot

location. – Calculate energy of filtered signal, compare to a threshold.

• Problems:– Pilot could be in a deep fade: quite common.– Threshold is susceptible to noise uncertainty.– Uncertainty in pilot location: could require a 100kHz bandwidth filter.

The larger the filter bandwidth, the worse the performance.

• Challenge: Can DTV pilot detection be made to provide robust sensing at SNR = -20 dB?

March 2007

Monisha Ghosh, Philips

Slide 3

doc.: IEEE 802.22-07/0125-00-0000

Submission

Alternate approaches

• Pilot-energy detection: can be made more robust by narrowing the filter-bandwidth.– Can this be done without compromising the detectability of signals

with large frequency offsets?

• Pilot location detection, rather than pilot energy.– More robust to thresholds: insensitive to noise uncertainty.– Performance is determined by pilot energy, however increasing

sensing time improves performance.

• Two methods proposed in this presentation:– FFT-based approach: pilot-energy and location detection.– Loop-based: pilot-location detection.

March 2007

Monisha Ghosh, Philips

Slide 4

doc.: IEEE 802.22-07/0125-00-0000

Submission

FFT-based pilot detection

• Demodulate signal to baseband, using nominal pilot position.

• Filter, using a low-pass filter. We used a BW of +/- 20 kHz.

• Subsample the signal. We reduced the sample rate from 21.52 MHz to 53.8 kHz (1/400).

• Take FFT of the subsampled signal. The FFT size will depend on the dwell-time. With parameters above:– 1 ms dwell will allow 32-point FFT.

– 5 ms dwell will allow 256-point-FFT.

• Pilot-energy detection: find maximum of FFT output-squared, and compare to a threshold. (NOT average over all FFT bins, but maximum).

• Pilot-location detection: compare location of maximum of FFT-output between multiple dwells.

March 2007

Monisha Ghosh, Philips

Slide 5

doc.: IEEE 802.22-07/0125-00-0000

Submission

Frequency Domain

0

x(t)

0

y(t)

53.8 kHz -53.8 kHz

……

21.52 MHz -21.52 MHz

……

March 2007

Monisha Ghosh, Philips

Slide 6

doc.: IEEE 802.22-07/0125-00-0000

Submission

Pilot Location Detection

• Let no. of dwells = N.• Let be the location of the maximum of the FFT-output

averaged over the first N/2 dwells.• Let be the location of the maximum of the FFT-output

averaged over the second N/2 dwells.• Detection statistic:

• If D < NT , signal present.

• Other variations:– Any PSD algorithm can be used instead of FFT.– Other averaging intervals could be used.

• PFA using this method is extremely low and robust. Threshold value NT will depend on the FFT-size.

)1(maxf

)2(maxf

)2(max

)1(max ffD

March 2007

Monisha Ghosh, Philips

Slide 7

doc.: IEEE 802.22-07/0125-00-0000

Submission

Signals From MSTV

Signal 1 WAS_3_27_06022000_REF.ASC

Signal 2 WAS_51_35_05242000_REF.ASC

Signal 3 WAS_311_36_06052000_ref.ASC

Signal 4 WAS_311_48_06052000_ref.ASC

Signal 5 WAS_68_36_05232000_ref.ASC

Signal 6 WAS_49_34_06142000_opt.ASC

Signal 7 WAS_311_35_06052000_ref.ASC

Signal 8 WAS_06_34_06092000_ref.ASC

Signal 9 WAS_49_39_06142000_opt.ASC

Signal 10 WAS_47_48_06132000_opt.ASC

Signal 11 WAS_32_48_06012000_OPT.ASC

Signal 12 WAS_86_48_07122000_ref.ASC

March 2007

Monisha Ghosh, Philips

Slide 8

doc.: IEEE 802.22-07/0125-00-0000

Submission

Performance with 1 ms dwell time, 10 dwells

32-point FFT

March 2007

Monisha Ghosh, Philips

Slide 9

doc.: IEEE 802.22-07/0125-00-0000

Submission

Performance with 5 ms dwell time, 6 dwells

256-point FFT

March 2007

Monisha Ghosh, Philips

Slide 10

doc.: IEEE 802.22-07/0125-00-0000

Submission

Performance with 5 ms dwell time, 10 dwells

256-point FFT

March 2007

Monisha Ghosh, Philips

Slide 11

doc.: IEEE 802.22-07/0125-00-0000

Submission

32-point FFT for Signal 7 (WAS_311_35_06052000)

Strong pilot, easily detected.

March 2007

Monisha Ghosh, Philips

Slide 12

doc.: IEEE 802.22-07/0125-00-0000

Submission

32-point FFT for Signal 9 (WAS_49_39_06142000)

Faded pilot, hard to detect

March 2007

Monisha Ghosh, Philips

Slide 13

doc.: IEEE 802.22-07/0125-00-0000

Submission

256-point FFT for Signal 9 (WAS_49_39_06142000)

Faded pilot,easily detected

March 2007

Monisha Ghosh, Philips

Slide 14

doc.: IEEE 802.22-07/0125-00-0000

Submission

Threshold sensitivity for signal energy detection

March 2007

Monisha Ghosh, Philips

Slide 15

doc.: IEEE 802.22-07/0125-00-0000

Submission

Threshold sensitivity for pilot energy detection

March 2007

Monisha Ghosh, Philips

Slide 16

doc.: IEEE 802.22-07/0125-00-0000

Submission

Threshold sensitivity for pilot location detection

March 2007

Monisha Ghosh, Philips

Slide 17

doc.: IEEE 802.22-07/0125-00-0000

Submission

Summary Of FFT-based detection

• SNR = -20 dB, pilot-energy detection: PD = 100% for all 12 channels, PFA = 2%, with ten 5 ms dwells.

• SNR = -20 dB, pilot-location detection: PD = 100% for 11 channels, and 90% for 1 channel, PFA = 3%, with ten 5 ms dwells.

• Thresholds for pilot-energy and pilot-location detection are less sensitive than threshold for signal-energy detection.

March 2007

Monisha Ghosh, Philips

Slide 18

doc.: IEEE 802.22-07/0125-00-0000

Submission

Loop-based pilot detection

• Two carrier recovery modules, each preset to a different frequency.

• The difference of the loop outputs of the two carrier recovery modules is compared against a threshold:– Signal is assumed to be present if the difference is less than the threshold.

• Not very sensitive to the threshold:– We used very conservative thresholds.

• Probability of false alarm is very low (close to 0) for a large range of threshold values.

• Sensing time can be reduced by adjusting the loop parameters.

March 2007

Monisha Ghosh, Philips

Slide 19

doc.: IEEE 802.22-07/0125-00-0000

Submission

Loop output behavior in the absence of DTV

• Not much variation from the initial frequency setting for the two loops.– Loops do not converge.

March 2007

Monisha Ghosh, Philips

Slide 20

doc.: IEEE 802.22-07/0125-00-0000

Submission

Loop output behavior in the presence of DTV

• Monotonic behavior in loop output in the presence of DTV signal for wide range of gain settings and SNRs.– Large threshold margin for reliable detection.

March 2007

Monisha Ghosh, Philips

Slide 21

doc.: IEEE 802.22-07/0125-00-0000

Submission

Performance after 75 ms, PFA << .01

March 2007

Monisha Ghosh, Philips

Slide 22

doc.: IEEE 802.22-07/0125-00-0000

Submission

Conclusions• Pilot detection is a robust means of sensing, down to -20 dB.

– Two methods presented: FFT-based and loop- based.

• Pilot-location detection is extremely robust against noise uncertainty.

• Pilot-energy detection, using the FFT based method, essentially reduces the bandwidth of the filter and hence performance is improved, even for severely faded pilots.

• Both pilot energy and pilot location detection can be used with multiple short dwell times to deliver robust sensing, thus avoiding the need for a single long quiet period.– QoS can be maintained for voice traffic.