Graph-based cooperative symbol decoding

In this paper, we discuss a communication network with a cooperative transmitter and relay. The receiver calculates the hard decisions of received symbols from the observations obtained from the relay and the transmitter. The calculation of log-likelihood ratio at the receiver requires marginalization over a joint density, which is known to be bear high complexity in general. Marginalization over the joint distribution function is required to calculate the log-likelihood ratio at the receiver. To mitigate this problem, we propose an iterative graph-based method. The main advantage of the proposed method, relative to the conventional, direct method, is that it allows flexible adaptation to different network topologies and cooperative protocols, and addition of structures like channel estimation and decoding mechanisms. The proposed method offers low complexity solutions for complicated network topologies where the optimal solution cannot be calculated due to complexity issues.