Throughput of Wireless Relay Networks with Interference Processing

Consider a wireless relay network, where all nodes except the source and destination act as relays. The problem of evaluating the capacity for a source-destination pair in this network and determining the corresponding optimal transmission strategy have attracted considerable research attention recently. However, even for a single relay, complete solutions are still unknown. A popular method to obtain partial results is to fix the relay strategy. In this paper, we consider a situation where relays adopt the decode-and-forward approach with the possibility of network coding. Also, the nodes can receive information from multiple transmitters simultaneously. This is accomplished by allowing physical layer interference processing at nodes. This processing yields increased sum rate when compared to interference avoidance or treating interference as noise. A linear optimization model is proposed to determine the maximum achievable throughput with interference processing. Numerical evaluation done on some example networks shows significant throughput gains.