An on-demand rate-aware joint power control and routing scheme

Since both transmission power and modulation method can be varied in Mobile Ad hoc Networks (MANETs), it is necessary to take power control and rate adaptation into account when dealing with routing problems. In this paper, we propose a Q-Learning-based Rate-aware Power-Controlled Routing scheme (QLRPCR) to address the joint optimization of power, rate and routing in an on-demand and distributed manner. An evaluation considering traffic load and node mobility shows that QLRPCR improves the average latency and the packet delivery ratio, while maintaining an acceptable overhead level compared with other Q-learning or power-controlled-routing protocols.