Balancing Tradeoffs for Energy-Efficient Routing in MANETs Based on Reinforcement Learning

This paper proposes an energy-efficient path selection algorithm which aims at balancing the contrasting objectives of maximizing network lifetime and minimizing energy consumption routing in mobile ad hoc networks (MANETs). The method is based on a reinforcement learning technique called the on-policy Monte Carlo (ONMC) method. Simulation results under high mobility environments reveal that variants of the proposed method can achieve the lowest long-term cost, which is a function that depicts the optimal tradeoff balance in the long run, when compared with existing tradeoff balancing schemes such as variants of the conditional max-min battery capacity routing (CMMBR) (Toh, C.-K., 2001) and the best minimum combined-cost routing algorithm (J.-H. Chang et al., 2004).