Project: IEEE P Working Group for Wireless Personal Area Networks (WPANs)

Transcription

1 Sep 9 doc.: IEEE Project: IEEE P8.5 Working Group for Wireless Personal Area Networks (WPANs) Title: [Common Coherent and Non-Coherent Modulation Proposal] Date Submitted: [-Sep-9] Source: [Michael Mc Laughlin] Company & Affiliation [Decawave Ltd.] Address [Digital Depot, Thomas Street, Dublin 8, Ireland] Voice:[ ], FAX: [What s a FAX?], E Mail: [michael.mclaughlin@decawave.com] Re: [8.5.6] Abstract: [Proposes using an 8.5.4a type FEC scheme for BAN ] Purpose: [To promote discussion in ] Notice: This document has been prepared to assist the IEEE P8.5. 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 P8.5. Slide

2 Sep 9 doc.: IEEE a Burst Position Modulation ~ns for 85kbps coherent and non-coherent code -code code -code g,g= (5,) for polarity and position respectively Slide

3 Sep 9 doc.: IEEE Antipodal 4pt Constellation bit bit Symbol Antipodal and bits different Hamming distance = and SED = d Transmitted Antipodal and bits different Hamming distance = and SED =d SED = Squared Euclidean Distance Orthogonal and bit different Slide 3

4 Sep 9 doc.: IEEE Non-Antipodal 4 pt Constellation Hamming distance = but SED = d bit bit Symbol Transmitted Orthogonal and bits different Antipodal and bit different Hamming distance = but SED =d SED = Squared Euclidean Distance Slide 4

5 Sep 9 doc.: IEEE Trellis Diagram, Optimum antipodal conv. code with K=5, g=35, g= d free = 7, with parallel pathways This is the way this code is supposed to work! Slide 5

6 Sep 9 doc.: IEEE Trellis Diagram, Non-Antipodal K=5, g=35, g= d free = 5, pathway This is how it actually works with the PPM/BPSK constellation! Slide 6

7 state Sep 9 doc.: IEEE Trellis Diagram, Non-Antipodal K=5, g=35, g= time d free = 7, parallel pathways This is a better code with the PPM/BPSK constellation! Slide 7

8 state Sep 9 doc.: IEEE Trellis Diagram, Non-Antipodal K=5, g=33, g= time d free = 7, parallel pathways This is just as good too and its systematic! Slide 8

9 PER Sep 9 doc.: IEEE Coherent vs Non-Coherent AWGN PER 5.4a PER, Coherent & Non-Coherent vs Best 8-State BPSK Convolutional Code Best 8-state BPSK Conv. Code 5.4a Coherent Conv Code 5.4a Non-coherent Eb/No (db) At Mbps required Eb/No of db means shorter non-coherent range. Range increases as bit rate decreases. 7m for Mbps => 5m for kbps Slide 9

10 Sep 9 doc.: IEEE a Convolution code XOR Pos n Sign Symbol Transmitted data in Note: Can recover Pos n bit with non-coherent receiver architecture. Exact code composition TBD Sign Pos n Coded data out Figure 7. Convolutional code. Rate ½ code, K=3, Generators (5,), g=, g= Slide

11 Sep 9 doc.: IEEE Summary Combined PPM and BPSK for both coherent and non-coherent modulation at all PRFs. K=3 convolutional code with octal generators (5,) for polarity and position respectively Small Viterbi Decoder ~3k gates <.mm in 9nm CMOS Slide

12 Sep 9 doc.: IEEE Advantages Non-coherent can decode also Don t need a common lower performance mode to accommodate non-coherent receiver No Price to Pay for Coherent Optimal AWGN BER performance for coherent architecture Coherent may turn off its Viterbi decoder systematic code makes this easy If signal good enough, (lose ~8dBs) lower power consumption Slide