Correlated Sources Transmission in Orthogonal Multiple Access Relay Channel: Theoretical Analysis and Performance Evaluation

In this paper, we consider the problem of transmitting two correlated binary sources over orthogonal multiple access relay channel (MARC), where two sources are communicating with a common destination with the assistance of a single relay. We assume decode-and-forward relaying strategy, and bit-wise exclusive or (XOR) network coding is performed at the relay node. First, a joint source-channel-network (JSCN) decoding technique is proposed to fully exploit the correlation between the sources, as well as the benefit of network coding. Then the achievable compression rate region of this system is derived based on the theorem for source coding with side information. It is found that the region is a 3-dimensional space surrounded by a polyhedron. Furthermore, the performance limit in Additive White Gaussian Noise (AWGN) channels and the outage probability in block Rayleigh fading channels are derived based on the achievable compression rate region. It is shown that the outage probability can be expressed by a set of triple integrals over the achievable compression rate region. The impact of source correlation on the performance of the system is investigated through asymptotic tendency analysis. The effectiveness of the proposed JSCN decoding technique and the accuracy of the theoretical analysis have been verified through a series of computer simulations, assuming practical channel codes. It is also shown that, as long as the source-relay links are perfect, the 2nd order diversity is always achieved with our proposed technique regardless of the strength of the source correlation.