1. doc.: IEEE 15-05-0xxx-00-004a Submission May 2005 Zafer Sahinoglu, Mitsubishi Electric Research Labs Slide 1 Project: IEEE P802.15 Working Group for Wireless Personal Area Networks (WPANs) Submission Title: [ Pulse Compression Techniques ] Date Submitted: [ May 2005 ] Source: [ Zafer Sahinoglu] Company: [Mitsubishi Electric ] E-Mail: [zafer@merl.com] Re: [Response to Call for Proposals] Abstract [This document explains pulse compression techniques that benefit precision ranging using IEEE 802.15.4a radios] Purpose:[Providing technical contributions for standardization by IEEE 802.15.4a in the ranging field] Notice: This document has been prepared to assist the IEEE P802.15. 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 acknowledges and accepts that this contribution becomes the property of IEEE and may be made publicly available by P802.15.

2. doc.: IEEE 15-05-0xxx-00-004a Submission May 2005 Zafer Sahinoglu, Mitsubishi Electric Research Labs Slide 2 Outline Definitions Existing Techniques Concluding Remarks

3. doc.: IEEE 15-05-0xxx-00-004a Submission May 2005 Zafer Sahinoglu, Mitsubishi Electric Research Labs Slide 3 Definitions Pulse Compression (or pulse coding) –A set of signal processing techniques to increase sensitivity and resolution in ranging systems Increasing the average power by having a longer pulse for better reception Changing bandwidth without changing pulse duration to have better range resolution –Range resolution depends on the bandwidth and SNR Integrated Side-lobe Level (ISL) –10log [total sidelobe power / peak response]

4. doc.: IEEE 15-05-0xxx-00-004a Submission May 2005 Zafer Sahinoglu, Mitsubishi Electric Research Labs Slide 4 Unmodulated vs. Modulated Pulse B: Bandwidth, T: Pulse duration –Unmodulated pulse => BT = 1 –Modulated pulse => BT > 1 By using spread spectrum techniques Unmodulated pulse Modulated pulse TpTp TpTp

5. doc.: IEEE 15-05-0xxx-00-004a Submission May 2005 Zafer Sahinoglu, Mitsubishi Electric Research Labs Slide 5 Power vs. Pulse Duration Energy content of the symbol would be the same (T 1 P 1 =T 2 P 2 ) –Long pulses with low power –Short pulses with high power

6. doc.: IEEE 15-05-0xxx-00-004a Submission May 2005 Zafer Sahinoglu, Mitsubishi Electric Research Labs Slide 6 Receivers Coherent receivers benefit from pulse compressions –Correlate the received signal with a replica of the transmitted modulated signal Non-coherent receivers see bulk energy with the same duration as unmodulated

7. doc.: IEEE 15-05-0xxx-00-004a Submission May 2005 Zafer Sahinoglu, Mitsubishi Electric Research Labs Slide 7 Pulse Compression Pros and Cons Pros Lower pulse power Higher maximum range Good range resolution Better interference rejection Cons Time side lobes Added transmitter and receiver complexity

8. doc.: IEEE 15-05-0xxx-00-004a Submission May 2005 Zafer Sahinoglu, Mitsubishi Electric Research Labs Slide 8 Pulse Compression Design Problem To find a waveform for coding of which –The autocorrelation function provides a strong peak with low side lobes –Better fits to FCC spectral mask

9. doc.: IEEE 15-05-0xxx-00-004a Submission May 2005 Zafer Sahinoglu, Mitsubishi Electric Research Labs Slide 9 Existing Pulse Compression Methods Binary phase coding (meteorological radars) –Repeatedly flipping the phase of the radio frequency signal within the duration of the pulse, according to a binary code (e.g., Barker codes) Poly-phase coding Frequency modulation Frequency stepping

10. doc.: IEEE 15-05-0xxx-00-004a Submission May 2005 Zafer Sahinoglu, Mitsubishi Electric Research Labs Slide 10 A Look Into Barker Codes Length 7 {+, +, +, -, -, +, -} Length 11 {+, +, +, -, -, -, +, -, -, +, -} Length 13 {+, +, +, +, +, -, -, +, +, -, +, -, +}

11. doc.: IEEE 15-05-0xxx-00-004a Submission May 2005 Zafer Sahinoglu, Mitsubishi Electric Research Labs Slide 11 A Look Into Barker Codes (2) Side lobe values less than or equal to 1/N –N: code length –Maximum output is normalized to 1

12. doc.: IEEE 15-05-0xxx-00-004a Submission May 2005 Zafer Sahinoglu, Mitsubishi Electric Research Labs Slide 12 Pseudo-random (PN) Codes Easy to generate Low but not minimal side lobes –For large N (>127), peak side lobe approximates to 1/N Maximal length PN codes (m-sequences) are the pretty efficient They have relatively high ISL –Unsuitable in highly clutter environments

13. doc.: IEEE 15-05-0xxx-00-004a Submission May 2005 Zafer Sahinoglu, Mitsubishi Electric Research Labs Slide 13 Poly-phase Codes Uses multiple phase shifts, as opposed to binary as in Barker Welty and Golay generates equal but different polarity side lobes –Total sidelobe cancellation can be achieved at the expense of two sets of matched filters, a long delay line and a summing function –Welty: {+, +, +, -, -, -, +, -} –Golay: {+, +, +, -, +, +, -, +}

14. doc.: IEEE 15-05-0xxx-00-004a Submission May 2005 Zafer Sahinoglu, Mitsubishi Electric Research Labs Slide 14 Suggestions To benefit coherent ranging a long coding sequence is needed. The sequence should be short enough to help non-coherent ranging