Paper Info
Title
Achievable Rates for the Three User Cooperative Multiple Access Channel
Abstract
For a three user Gaussian multiple access channel (MAC), we propose a new superposition block Markov encoding based cooperation scheme. Our scheme allows the three users to simultaneously cooperate both in pairs, and collectively, by dividing the transmitted messages into sub-messages intended for each cooperating partner. The proposed encoding and decoding at the transmitters take into account the relative qualities of the cooperation links between the transmitters. We obtain and evaluate the achievable rate region based on our encoding strategy, and compare it with the achievable rates for the two user cooperative MAC. We demonstrate that the added diversity by the presence of the third user improves the region of achievable rates, and this improvement is especially significant as far as the sum rate of the system is concerned. I. INTRODUCTION The growing demand for high data rate mobile applica- tions challenges researchers to develop wireless communication systems which are able to accommodate a higher number of concurrent users, communicating reliably at improved rates. Although an increase in the number of users in a system seems to cause more interference and hence worse performance, this interference may actually be viewed as free side information, which is distributed to all communicating parties thanks to the propagative nature of the wireless communication channel. Therefore, if the users are allowed to make use of the free side information and cooperate in sending each other's messages, the diversity provided to the participating users will increase with increasing number of users, potentially leading to higher rates. The idea of user cooperation diversity in wireless networks can be traced back to the pioneering works of Carleial (1) and Willems et al. (2) on multiple access channels with generalized feedback (MAC-GF). The MAC-GFs in (1) and (2) are both modelled by (X1 ×X2 ,P (y, y1 ,y 2|x1 ,x 2), Y×Y 1 ×Y2), where the receiver observes the channel output Y , and the transmitters have access to separate channel outputs Y1 and Y2 through feedback links. The transmitters are then able to encode their messages based on these feedback signals, and can achieve higher rates than the traditional MAC. Carleial and Willems et al. used different versions of block Markov superposition encoding, as well as different decoding policies (sliding win- dow decoding versus backwards decoding respectively), to obtain achievable rate regions for the MAC-GF. The rate region This work was supported by The Scientific & Technological Research
Year
DOI
Venue
2008
10.1109/WCNC.2008.270
WCNC
Keywords
Field
DocType
Gaussian channels,Markov processes,block codes,decoding,multi-access systems,multiuser channels,decoding,multiple access channel,superposition block Markov encoding based cooperation scheme,user cooperative Gaussian MAC
Superposition principle,Markov process,Computer science,Markov chain,Block code,Computer network,Communication channel,Gaussian,Decoding methods,Encoding (memory)
Conference
ISSN
Citations 
PageRank 
1525-3511
8
0.63
References 
Authors
7
2
Name
Order
Citations
PageRank
Çagatay Edemen1121.08
Onur Kaya210613.85