Channel allocation in a cognitive radio network using non deterministic Q learning algorithm

Cognitive radio technology has been emerged to provide the solution of improvement of spectrum utilization. In this paper we try to solve the problem of allocation of channels to the secondary users using non deterministic Q learning technique. Using Q learning technique, we try to provide prior knowledge to the secondary users about the channel usage pattern of licensed users so that a secondary user can efficiently select channels in a multi cell cognitive radio network. Firstly, a channel assignment algorithm based on non deterministic Q learning technique has been proposed in this paper. Using ageing technique, starvation of low priority users is removed from the algorithm. Simulation result shows that the proposed channel assignment algorithm is effective.