March 2005 doc IEEE 802 22 050010 r

  • Slides: 13
Download presentation
March 2005 doc. : IEEE 802. 22 -05/0010 r 0 DTV Signal Spectra and

March 2005 doc. : IEEE 802. 22 -05/0010 r 0 DTV Signal Spectra and Filtering IEEE P 802. 22 Wireless RANs Date: 2005 -02 -25 Author: 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] org. > Submission 1 Paul Thompson, Paul Thompson Associates, LLC

March 2005 doc. : IEEE 802. 22 -05/0010 r 0 Abstract As an aid

March 2005 doc. : IEEE 802. 22 -05/0010 r 0 Abstract As an aid to visualizing how an ATSC Pilot signal could be used to detect a DTV station at a signal level below that which enables usable DTV service, a local DTV station was extracted off the air and subjected to varying amounts of pre- and post-detection filtering by a spectrum analyzer. Images of the resulting spectra are presented for reference in IEEE 802. 22 discussions. Submission 2 Paul Thompson, Paul Thompson Associates, LLC

March 2005 doc. : IEEE 802. 22 -05/0010 r 0 DTV Signal Spectra and

March 2005 doc. : IEEE 802. 22 -05/0010 r 0 DTV Signal Spectra and Filtering While it is expected that any cognizant receiver employed to detect ATSC DTV signals in satisfaction of FCC Part 15. 244 will have to be somewhat sophisticated in its feature extraction*, it is nevertheless instructive to understand the capabilities of a simple DTV Pilot power (square-law) detector. The ATSC DTV Pilot is 11. 3 d. B below the total average power of the DTV signal in a 6 MHz bandwidth. As long as the Pilot signal is sufficiently specular (i. e. , it does not have excessive phase noise or other modulation which broadens its bandwidth), it theoretically should be possible to detect it with a pre-detection bandwidth much smaller than 6 MHz and thus improve the sensitivity of the detection. Additional post-detection filtering of the Pilot signal should theoretically be able to improve the sensitivity of the Pilot detection even further. As an experiment, I fabricated the simplest dipole antenna I could construct (Figure 1, Slide 4) and connected it directly to a spectrum analyzer. No preamplifier was used between the antenna and the spectrum analyzer, so the noise figure of the Spectrum analyzer receiving system is at least 35 d. B in the TV bands. The antenna was placed where there was line of sight to, and a few miles from an ATSC DTV station transmitting tower (KVAL in Eugene, Oregon, Channel 25. ) The antenna was copolarized to the DTV signal and oriented with maximum gain toward the DTV transmitter. The spectrum analyzer was tuned to a center frequency of 539 MHz, a Span of 8 MHz, a pre-detection bandwidth of 3 KHz and a post-detection (video) bandwidth of 3 KHz. The resulting spectrum is shown in Figure 2, Slide 5. _________ *For instance, the Comments to the FCC NPRM from SSC (Ref 1) described that a FCC Part-15. 244 -compliant cognizant receiver must discriminate against spurious signals from other electronic equipment in proximity to the radio. (continued on Slide 6) Submission 3 Paul Thompson, Paul Thompson Associates, LLC

March 2005 doc. : IEEE 802. 22 -05/0010 r 0 Figure 1 - Dipole

March 2005 doc. : IEEE 802. 22 -05/0010 r 0 Figure 1 - Dipole antenna and Spectrum Analyzer Submission 4 Paul Thompson, Paul Thompson Associates, LLC

March 2005 doc. : IEEE 802. 22 -05/0010 r 0 Figure 2 - Line-of-Sight

March 2005 doc. : IEEE 802. 22 -05/0010 r 0 Figure 2 - Line-of-Sight KVAL DTV Siganl Submission 5 Paul Thompson, Paul Thompson Associates, LLC

March 2005 doc. : IEEE 802. 22 -05/0010 r 0 DTV Signal Spectra and

March 2005 doc. : IEEE 802. 22 -05/0010 r 0 DTV Signal Spectra and Filtering (continued) The spectrum displayed in Figure 2 is characterized by a relatively flat power density over the 5. 38 MHz of occupied bandwidth with the Pilot signal clearly observable on the lower end of the spectrum. Also, notice that the intrinsic noise of the spectrum analyzer, observable at the extreme right and left-hand regions of the image, is at least 30 d. B below the main signal level in the vicinity of the Pilot. In the discussion below, it will be important to know that the spectrum analyzer noise is insignificant in the region of the Pilot signal. The choice of a pre-detection bandwidth of 3 KHz was made because it is roughly equivalent to using a 2048 point FFT to analyze the 6 MHz-wide Channel. The 2048 point FFT is a typical one which might be used to implement an OFDM/OFDMA radio and thus might already be present in a FCC Part 15. 244 -compliant radio. In an effort to explore in more detail the nature of the detected Pilot, the spectrum analyzer Span was reduced to 100 KHz and the center frequency changed to be that of the Pilot signal. The 3 KHz pre-detection and post-detection bandwidths were retained. The resulting display is shown in Figure 3, Slide 7. (continued on Slide 8) Submission 6 Paul Thompson, Paul Thompson Associates, LLC

March 2005 doc. : IEEE 802. 22 -05/0010 r 0 Figure 3 - DTV

March 2005 doc. : IEEE 802. 22 -05/0010 r 0 Figure 3 - DTV Pilot Signal Submission 7 Paul Thompson, Paul Thompson Associates, LLC

March 2005 doc. : IEEE 802. 22 -05/0010 r 0 The noise-like component of

March 2005 doc. : IEEE 802. 22 -05/0010 r 0 The noise-like component of the spectrum in Figure 3 is, of course, due to the data modulation of the DTV carrier, not spectrum analyzer noise. The Pilot-to-DTV Noise ratio in the region of the Pilot signal is approximately +25 d. B as would be expected when using the 3 KHz pre-detection bandwidth: +33 d. B of bandwidth improvement 11. 3 d. B of Pilot suppression +3 d. B. The last 3 d. B is because the Pilot signal is located in a region where the power density of the main DTV modulation is approximately 3 d. B below its maximum across the 5. 38 MHz of occupied spectrum. If one were to visualize a spectral-contrast receiver attempting to detect the DTV Pilot signal of Figure 3, one could take noise samples from one or more adjacent frequency bins, normalize their power and use this value to set a threshold for the detection of the Pilot. With a 3 KHz pre-and post detection bandwidth, this process could occur approximately every 1/3 millisecond so as to get independent samples. Since the entire point-to-multipoint network must be silent during the search for the pilot, creating a 1/3 millisecond cessation in the transmissions from all stations is not onerous: on the order of the time required for the round-trip delay to/from a 25 -mile distant terminal. Of course, the Noise that limits detection of the Pilot signal will not always be that of the DTV modulation itself. For instance, a receiver with a 0 d. B gain, co-polarized antenna and located at the edge of the Grade B Contour will measure -90 d. Bm (in the 6 MHz bandwidth) from a line-of sight DTV Station. If that same receiver has a Noise Figure of 6 d. B, the power spectral density of the DTV signal modulation will be 10 d. B above the power spectral density of the receiver noise. In this unobstructed case, the ability to detect the Pilot will be limited by the DTV signal modulation noise as is observed in Figure 3. If the DTV signal experiences 10 d. B of obstruction or if the receiver is 10 d. B in range farther from the transmitter, the receiver and DTV modulation power densities will be equal. For further reductions in the DTV signal power, the receiver noise will limit the ability to detect the Pilot. (continued on Slide 9) Submission 8 Paul Thompson, Paul Thompson Associates, LLC

March 2005 doc. : IEEE 802. 22 -05/0010 r 0 To explore the effect

March 2005 doc. : IEEE 802. 22 -05/0010 r 0 To explore the effect of post-detection filtering, the post-detection (video) filter of the spectrum analyzer was varied to produce the spectra in the center column of Figure 4, Slide 10. Figure 4 (d) is a duplicate of Figure 3 (3 KHz preand post-detection filters. ) Figures 4 (e) and (f) show a 3 KHz pre-detection filter with post-detection filters of 1 KHz and 100 Hz, respectively. The variance of the noise and the Pilot+noise components determines where the Pilot detection threshold can be set to achieve a specific Probability of Detection concurrent with maintaining an acceptable False Alarm Rate. It can be seen that the variance of the noise and Pilot+noise components is reduced as the post-detection filtering bandwidth is reduced. Reducing the variance of the noise and Pilot+noise components translates into being able to reduce the Pilot detection threshold (at a constant False Alarm Rate) thus enabling detection at a lower Pilot power (at a constant probability of Detection. ) Reducing the spectrum analyzer post-detection (video) bandwidth is equivalent to taking a number of successive noise samples and averaging them. It should be kept in mind that the narrower the post-detection bandwidth, the longer it will take to obtain a valid sample. For instance, with a 100 Hz postdetection bandwidth (or equivalent averaging), it will take on the order of 10 milliseconds to obtain a valid sample. This time may be too long for routine confirmation that an occupied RF channel does not have a DTV signal. This suggests that the Pilot detection receiver might have a variety of post-detection bandwidths at its disposal to address initial scanning of the entire RF environment (where more time might be afforded) or periodic confirmation of an occupied channel (where long interruptions of the Point-to-multipoint network cannot be tolerated. ) (continued on Slide 11) Submission 9 Paul Thompson, Paul Thompson Associates, LLC

March 2005 doc. : IEEE 802. 22 -05/0010 r 0 (c) Pre: 10 KHz

March 2005 doc. : IEEE 802. 22 -05/0010 r 0 (c) Pre: 10 KHz Post: 100 Hz (f) Pre: 3 KHz Post: 100 Hz (b) Pre: 10 KHz Post: 1 k. Hz (e) Pre: 3 KHz Post: 1 k. Hz (a) Pre: 10 KHz Post: 10 k. Hz (h) Pre: 1 KHz Post: 100 Hz (g) Pre: 1 KHz Post: 1 k. Hz (d) Pre: 3 KHz Post: 3 k. Hz Figure 4 - Effect of Pre- and Post Detection Bandwidths Submission 10 Paul Thompson, Paul Thompson Associates, LLC

March 2005 doc. : IEEE 802. 22 -05/0010 r 0 For completeness, the spectrum

March 2005 doc. : IEEE 802. 22 -05/0010 r 0 For completeness, the spectrum analyzer pre-detection bandwidth was increased to 10 KHz and reduced to 1 KHz to produce the spectra in Figure 4 (a-c) and (g-h), respectively. A 10 KHz pre-detection bandwidth would roughly correspond to a 512 point FFT analyzing a 6 MHz channel. Likewise, a 1 KHz pre-detection would roughly correspond to a 8196 -point FFT analyzing the same channel. The post-detection bandwidths are shown in the figure captions. The ability to get a clean spectrum with a 1 KHz bandwidth suggests, at least for this DTV station, that the Pilot signal is very specular and could legitimately be detected by 1 KHz bandwidths, or smaller (if the sample time can be afforded. ) The spectrum shown in Figure 2 was taken under ideal, line-of-sight conditions. As an exercise, the antenna and spectrum analyzer were brought indoors and the spectrum of the same DTV station obtained as shown in Figure 5. Because of dispersion present in the indoor signal, the Pilot portion of the spectrum was substantially reduced whereas the rest of the DTV signal is relatively at full power. Admittedly, I positioned the antenna to produced this effect. However, it does demonstrate that selective fading of the Pilot signal with respect to the total ATSC signal could be a factor in detecting DTV signals if reliance was placed on only detecting the Pilot. Submission 11 Paul Thompson, Paul Thompson Associates, LLC

March 2005 doc. : IEEE 802. 22 -05/0010 r 0 Figure 5 - Indoor

March 2005 doc. : IEEE 802. 22 -05/0010 r 0 Figure 5 - Indoor KVAL DTV Siganl Submission 12 Paul Thompson, Paul Thompson Associates, LLC

March 2005 doc. : IEEE 802. 22 -05/0010 r 0 References Reference 1: Appendix

March 2005 doc. : IEEE 802. 22 -05/0010 r 0 References Reference 1: Appendix to Reply Comment entitled “High Sensitivity Broadcast TV Signal Detection” submitted by Spectrum Sharing Company to FCC NPRM for FCC Part 15. 244, dated January 21, 2005 (available from FCC Electronic Comment Filing System web site. ) Submission 13 Paul Thompson, Paul Thompson Associates, LLC