Cooperative Retransmissions Using Markov Decision Process With Reinforcement Learning

In cooperative retransmissions, nodes with better channel qualities help other nodes in retransmitting a failed packet to its intended destination. In this paper, the authors propose a cooperative retransmission scheme where each node makes local decision to cooperate or not to cooperate at what transmission power using a Markov decision process with reinforcement learning. With the reinforcement learning, the proposed scheme avoids solving a Markov decision process with a large number of states. Through simulations, they show that the proposed scheme is robust to collisions, is scalable with regard to the network size, and can provide significant cooperative diversity.