Cooperative Communication In Wireless Networks

1
Project IEEE P802.15 Working Group for Wireless
Personal Area Networks (WPANs) Submission Title
Ranging Signal Waveforms Non-coherent Ranging
Proposals (Overview) Date Submitted 13 June,
2005 Source Zafer Sahinoglu Company
Mitsubishi Electric. Address 201 Broadway,
Cambridge, MA 02139 Voice617 621-7588, FAX
617 621-7549, E-Mail zafer_at_merl.com Re
802.15.4a. Abstract Overview of ranging
receiver architectures for non-coherent
reception Purpose To promote discussion in
802.15.4a. 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
Outline

• Signal Waveforms
• Receiver Architecture
• Comparison of Proposals
• Issues in Non-coherent ranging
• Suggestions

3
Signal Waveforms

• There are 4 waveforms under consideration
• Bulk PPM
• IR-TR (Impulse Radio Transmitted Reference)
• MTOK sequences
• TH-IR (Time Hopping Impulse Radio)

4
Option-I
One Bit
Always Empty
Always Empty
Always Empty
100ns
8-chip times 150ns
100ns
8-chip times 150ns
The Other Bit
Always Empty
Always Empty
Always Empty
100ns
8-chip times 150ns
Enough long not to cause IFI 100ns
8-chip times 150ns
5
Option-II
 11 
2-PPM TR base M 2 One bit/symbol
 01 
 10 
 00 
(coherent decoding possible)
6
To compare fairly, with option-I, which has 8
pulses in 500ns, the PRI for option-III is 62.5ns
Option-III
Ternary Signaling for Synchronization Ranging
-Common signaling (Mode 1) for ALL
Detectors -Receiver-specific signaling (Mode 2)
for ED
Pulse Repetition Interval 62.5ns
1
2
3
31
4
5
6
7
8
30

Non-inverted pulses are blue, Inverted pulses are
green.
Synchronisation / Ranging preamble Binary Base
Sequence repeated For K times


.................
Symbol Interval 1937.5ns
Symbol Interval 1937.5ns
7
Option-IV

• Illustration
• Use of an 8-ary Time Hopping code of length 4
• Use of such a TH code combined with the band plan
  may allow to handle the SOP issue
• Code order and length are scalable to meet
  different requirements
• Tp 4ns, Tc 20 ns, Tf 160 ns, Tsymbol 640
  ns

PRP TH
Tp
Tc
Tf
8
Recommended Architecture for Ranging with
Non-Coherent Rx (MERL)

• (No FFT routine is needed, being different from
  doc0269)

Energy image generation
Removes interference
2-4ns
Length-3 Vertical Median or Minimum Filtering
1D to 2D Conversion
LPF / integrator
BPF
( )2
ADC
2D to 1D Conversion with Energy Combining
TOA Estimator
9
Recommended Architecture for Ranging with
Non-Coherent Rx (I2R)
TOA Estimator
Energy image generation interference suppression
2D-1D conver-sion
interference suppression
2-4ns
Energy combining across symbols
Sliding Correlator
1D-2D conversion
LPF / integrator
BPF
( )2
ADC
Bipolar Template
10
Recommended Architecture for Ranging with
Non-Coherent Rx (FT RD)
Time base 1-2ns accurate
4ns
Time-stamping
"Path-arrival dates" table
Analog Comparator
LPF / integrator
BPF
( )2
1D to 2D Conversion
Filtering Assumption/path selection
Assumption path synchronization Matrix
TOA measure
11
Commonalities

• Signal energy collection architecture is the same
  
• Sampling resolution is the same (e.g., 2ns-4ns)
• 1D to 2D conversion for interference removal
  seems to be adopted in all the architectures

Energy image generation
TOA Estimator
1D to 2D Conversion
Interference removal
2D to 1D Conversion
12
Ranging Architecture Comparison

• MERL architecture works with all signal waveforms
• Its performance would be
• Very high in Option-4 (FT)
• High in Option-1 (MERL, TDC)
• Moderate in Option-3 (I2R)
• I2R and MERL receiver architectures have good
  support for scalability in TDOA applications
  (unsure about FT)
• Receive from TDOA stations simultaneously,
  without changing the RF front
• Decode TOA from each station at the post filtering

Post Filtering
Energy Matrix Generator-M
2D-1D conversion
TOA Est.
13
Feature Comparison
Pulse OOK (option-III) Burst PPM (option-I) TH-IR
Signaling Spaced out pulse seq Clustered pulse seq Time Hopping Impulse Radio
Energy Integration period (for ranging) 24ns 24ns 2ns-4ns
Performance _at_1Mbps without SOP TBD TBD TBD
Performance _at_1Mbps with SOP Good in com. Mode Questionable in ranging (Sufficient processing gain to handle) good in ranging mode (Not much processing gain to handle the comm. mode) Good in ranging and comm. Pulse amplitudes may be a drawback
Common signaling for preamble No Time hopping Time Hopping Time Hopping
Additional complexity for Coherent receiver to receive preamble with common signaling No Yes (2 layer sync, TH then code de-spreading) TH
Inter-pulse interference during ranging operation Less due to long PRI (30ns _at_ 33MHz PRF or 60ns _at_ 16MHz PRF) More due to small inter pulse interval Less due to long PRI
Sampling rate in ranging Low in transmitter High in receiver Low in receiver High in transmitter High in receiver Low in transmitter
14
Issues in Non-coherent ranging

• Square-law operation is non-linear
• Relative improvement in SNR by statistical
  multiplexing is higher, when pulse amplitudes are
  higher
• Option-1 requires longer preamble than option-4
  to achieve the same SNR level
• Error floor can be lowered for the same SNR
  level, by having narrower integration intervals

-
Preamble length
For a given sampling resolution
Mean absolute ranging error
Higher sampling resolution

EbN0
25dB
15
Suggestions

• Test the ranging performance of the three
  proposals at a sufficiently high SNR without SOP
• To see the best achievable ranging accuracy with
  90 confidence level
• To derive the required preamble length
• Test the ranging performance of the three
  proposals at the 90 level SNR from step-1 with
  SOP (SIR0dB, 10dB)
• Quantify the degradation in ranging accuracy and
  confidence level