Cooperative communication is a promising solution for Wireless Sensor Networks (WSN). Its main idea consists in a source node (sensor) which uses, at each hop, the resources of multiple nodes (called cooperative nodes) to transmit its data. Thus, by sharing resources between nodes, the transmission quality is enhanced. In this paper, we use the technique of reinforcement learning by opponent modelling, and propose a cooperative communication protocol based on RSSI and node energy consumption in a competetive context (RSSI/Energy CC). For this algorithm, a competitive mechanism is implemented at each node using a multi-agent reinforcement learning algorithm. The reinforcement learning concept consists in considering the cooperative nodes as multiple agents learning their optimal policy through experiences and rewards. By adopting an optimal policy, agents can optimally learn using locally observed network information and limited information exchange. Simulation results show that the proposed algorithm performs well in terms of network lifetime, packet delay, and energy consumption.¹