An adaptive delay-based power control and routing scheme

In this paper, We propose a Q-learning-based routing scheme embedded with a power-control method to minimize the overall network delay performance. Various problems of applying reinforcement learning method in Mobile Ad hoc Networks (MANETs), such as optimization, the routing-loop problem and the convergence problem, are all addressed accordingly. A power-control scheme which makes use of the on-demand routing information to form an objective function are introduced in order to find an optimal power level that provides a suitable trade-off between transmission range and interference. The impact between power control and routing are also discussed in this paper. Simulations show that our method can achieve better performance under different load and mobility conditions.