Optimizing network-coded cooperative communications via joint session grouping and relay node selection

Network-coded cooperative communications (NC-CC) is a new paradigm in wireless networks that employs network coding (NC) to improve the performance of CC. The core mechanism to harness the benefits of NC-CC is to appropriately combine sessions into separate groups, and then have each group select the most beneficial relay node for NC-CC. In this paper, we study this joint grouping and relay node selection problem for NC-CC. Due to NP-hardness of problem, we propose a distributed and online algorithm that offers near-optimal solution to this problem. The key idea in our algorithm is to have each neighboring relay node of a new session determine and offer its best local group; and then to have the source node of the new session select the best group among all offers. We show that our distributed algorithm has polynomial complexity. Using extensive numerical results, we show that our distributed algorithm adapts well to online network dynamics.