1. September, 2005 Doc: IEEE 15-05-0524-00-004a Qi, Li, Umeda, Hara and Kohno (NICT) SlideTG4a1 Project: IEEE P802.15 Working Group for Wireless Personal Area Networks (WPANs) Submission Title: [Three ranging-related schemes] Date Submitted: [September, 2005] Source: [Yihong Qi, Huan-Bang Li, Masataka Umeda, Shinsuke Hara and Ryuji Kohno, Company: National Institute of Information and Communications Technology ] Contact: Yihong Qi Voice:+81 46 847 5092, E-Mail: yhqi@nict.go.jp] Abstract: [Three ranging-related schemes are presented: 1. for the problem that the first arriving signals are often weak and NLOS, positioning using mulitpath delays will improve the accuracy. 2. a reduced dimensional approach is proposed for the bad GDOP problem. 3. a coherent delay estimation scheme is devised which works well with low sampling rate and feasible ADC implementation.] Purpose: [to discuss three ranging-related schemes ] 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. September, 2005 Doc: IEEE 15-05-0524-00-004a Qi, Li, Umeda, Hara and Kohno (NICT) SlideTG4a2 Outline  Positioning using multipath delays (cf. first arrival detection)  Positioning in an ill-conditioned geometry (bad GDOP (geometric dilution of precision))  A coherent delay estimation scheme with low sampling rate  Conclusions

3. September, 2005 Doc: IEEE 15-05-0524-00-004a Qi, Li, Umeda, Hara and Kohno (NICT) SlideTG4a3 Two Positioning Schemes

4. September, 2005 Doc: IEEE 15-05-0524-00-004a Qi, Li, Umeda, Hara and Kohno (NICT) SlideTG4a4 Current/conventional schemes  Ranging: first arrival detection  Positioning: based on multiple range estimates triangulation weighted least square (LS) methods

5. September, 2005 Doc: IEEE 15-05-0524-00-004a Qi, Li, Umeda, Hara and Kohno (NICT) SlideTG4a5 What are problems with the current schemes? Positioning accuracy will be degraded due to  Weak first arriving signals, e.g., 6dB lower than the strongest path.  NLOS first arriving signals  Bad GDOP (geometric dilution of precision)

6. September, 2005 Doc: IEEE 15-05-0524-00-004a Qi, Li, Umeda, Hara and Kohno (NICT) SlideTG4a6 Positioning using multipath delays

7. September, 2005 Doc: IEEE 15-05-0524-00-004a Qi, Li, Umeda, Hara and Kohno (NICT) SlideTG4a7 Motivation The second and later arriving signals also carry information on the position of interest. cf. weak and/or NLOS first arriving signals  Positioning using both  Multipath delays  Their statistic information (e.g., mean, variance)

8. September, 2005 Doc: IEEE 15-05-0524-00-004a Qi, Li, Umeda, Hara and Kohno (NICT) SlideTG4a8 Two numerical examples based on analytical results For illustration purpose, some simplifying assumptions on multipath delays:  Exponential or equal gain models  The minimum delay resolution being the inverse of chip duration  Gaussian NLOS delay variables

9. September, 2005 Doc: IEEE 15-05-0524-00-004a Qi, Li, Umeda, Hara and Kohno (NICT) SlideTG4a9 Numerical example 1 Positioning accuracy vs. num of multipath Equal gain Exponential gain with -6dB Exponential gain with -3dB Observation: use of more strong multipaths can improve the positioning accuracy

10. September, 2005 Doc: IEEE 15-05-0524-00-004a Qi, Li, Umeda, Hara and Kohno (NICT) SlideTG4a10 Numerical example 2 For a fair comparison: Using fixed total energy; relative accuracy improvement, compared with the conventional method using only the first arrivals Three types of system channels 1 2 3

11. September, 2005 Doc: IEEE 15-05-0524-00-004a Qi, Li, Umeda, Hara and Kohno (NICT) SlideTG4a11 Numerical example 2 (cont’d) relative accuracy improvement vs. standard deviation of NLOS delays Observation: using more multipaths is especially effective for accuracy improvement in wideband systems 1MHz 100MHz 5MHz

12. September, 2005 Doc: IEEE 15-05-0524-00-004a Qi, Li, Umeda, Hara and Kohno (NICT) SlideTG4a12 A reduced-dimensional method for bad GDOP (geometric dilution of precision) cases

13. September, 2005 Doc: IEEE 15-05-0524-00-004a Qi, Li, Umeda, Hara and Kohno (NICT) SlideTG4a13 What is the bad GDOP? Good GDOP case: nodes are distributed evenly a2 a3 a1 Mobile node a2 a3 a1 Mobile node Bad GDOP case: all nodes are lined up The error is small. The error is large.

14. September, 2005 Doc: IEEE 15-05-0524-00-004a Qi, Li, Umeda, Hara and Kohno (NICT) SlideTG4a14 What is the problem? a2 a3a3 a1 m Bad dim: x Good dim: y Two dimensional positioning estimation (x,y) vs. an essentially one-dimensional problem (y axis only)

15. September, 2005 Doc: IEEE 15-05-0524-00-004a Qi, Li, Umeda, Hara and Kohno (NICT) SlideTG4a15 A reduced dimension approach 1. Find the good dim(s) 2. Perform a regular positioning in the good dimension 3. Estimate the coordinate in the bad dim(s) separately

16. September, 2005 Doc: IEEE 15-05-0524-00-004a Qi, Li, Umeda, Hara and Kohno (NICT) SlideTG4a16 A simulation result for 2-D bad GDOP Conventional method Reduced dimensional method Theoretical limit Positioning accuracy vs. standard deviation of ranging errors

17. September, 2005 Doc: IEEE 15-05-0524-00-004a Qi, Li, Umeda, Hara and Kohno (NICT) SlideTG4a17 Flashback  Positioning using multipath delays For the problem of weak and/or NLOS first arriving signals Con: increased computation complexity  A reduced-dimensional approach for positioning For bad GDOP geometry

18. September, 2005 Doc: IEEE 15-05-0524-00-004a Qi, Li, Umeda, Hara and Kohno (NICT) SlideTG4a18 Coherent delay estimation with low sampling rate and feasible ADC implementation

19. September, 2005 Doc: IEEE 15-05-0524-00-004a Qi, Li, Umeda, Hara and Kohno (NICT) SlideTG4a19 CorrelatorA/D Delay estimation/ First-arrival detection A delay estimate A transmit signal A basic system model

20. September, 2005 Doc: IEEE 15-05-0524-00-004a Qi, Li, Umeda, Hara and Kohno (NICT) SlideTG4a20 Two ways of implementing ADC LO LPF code- correlator BPF π/2 ADC output Matched to Gaussian pulse Spreading code Difficult to implement easy to implement

21. September, 2005 Doc: IEEE 15-05-0524-00-004a Qi, Li, Umeda, Hara and Kohno (NICT) SlideTG4a21 What is the problem? t m+1 tn h(tn) h(tm) h(tm+1) h(tm+Z-1) t m+2 t m+Z Given samples of a correlation function, how to estimate the time instant corresponding to the peak? correlation function

22. September, 2005 Doc: IEEE 15-05-0524-00-004a Qi, Li, Umeda, Hara and Kohno (NICT) SlideTG4a22 What is information we know? t m+1 tn t m+2 t m+Z correlation autocorrelation correlation = autocorrelation of s(t) +noise The expression is known.Statistics is known. correlation function

23. September, 2005 Doc: IEEE 15-05-0524-00-004a Qi, Li, Umeda, Hara and Kohno (NICT) SlideTG4a23 A natural way to use all information Formulate maximum likelihood estimation (ML). However, it is complicated:  One dimension iterative searching  Nonlinear autocorrelation function involved  Lots of samples (N) involved

24. September, 2005 Doc: IEEE 15-05-0524-00-004a Qi, Li, Umeda, Hara and Kohno (NICT) SlideTG4a24 Our approach: simplified MLE t m+1 tn h(tn) h(tm) h(tm+1) h(tm+Z-1) t m+2 t m+Z Intuition: samples near the peak are more important.  Use less samples Taylor expansion of autocorrelation function around the peak

25. September, 2005 Doc: IEEE 15-05-0524-00-004a Qi, Li, Umeda, Hara and Kohno (NICT) SlideTG4a25 A simple solution where

26. September, 2005 Doc: IEEE 15-05-0524-00-004a Qi, Li, Umeda, Hara and Kohno (NICT) SlideTG4a26 A simple solution An algebraic solution, no iterative search Less than 4 samples in general No nonlinear function any more Independent of noise level Optimal in the sense that the estimate is approaching to the theoretical lower limit as over- sampling is sufficiently large.

27. September, 2005 Doc: IEEE 15-05-0524-00-004a Qi, Li, Umeda, Hara and Kohno (NICT) SlideTG4a27 Simulation parameters PRF=30.875MHz Sampling rate f s (ADC)=494MHz (=16xPRF) Ternary sequence with length of 31 Gaussian Pulse with bandwidth 500MHz AWGN Channel Conventional method: Pick up the largest sample Interpolation method: Not include the autocorrelation info.

28. September, 2005 Doc: IEEE 15-05-0524-00-004a Qi, Li, Umeda, Hara and Kohno (NICT) SlideTG4a28 Simulation result 1 RMS Estimation Error [nsec] E b /N 0 [dB] ADC after Code Correlator ADC before Code Correlator Conventional method Simplified ML Interpolation

29. September, 2005 Doc: IEEE 15-05-0524-00-004a Qi, Li, Umeda, Hara and Kohno (NICT) SlideTG4a29 Simulation result 2 E b /N 0 =-3dB Conventional method Interpolation Simplified ML

30. September, 2005 Doc: IEEE 15-05-0524-00-004a Qi, Li, Umeda, Hara and Kohno (NICT) SlideTG4a30 Advantages  Working well at low sampling rate (less than signal bandwidth)  Feasible ADC implementation  Low computation complexity Same level of complexity compared with conventional schemes  Independent of noise level Ongoing work: incorporating decay patterns for multipath scenarios

31. September, 2005 Doc: IEEE 15-05-0524-00-004a Qi, Li, Umeda, Hara and Kohno (NICT) SlideTG4a31 Conclusions  Positioning using multipath delays  A reduced dimensional approach for positioning in bad GDOP  A coherent delay estimation scheme with low sampling rate and feasible ADC implemetation