Joint Admission Control and Channel Selection Based on Multi Response Learning Automata (MRLA) in Cognitive Radio Networks

Abstract

We use multi response learning automata (MRLA) to control how secondary users should access the licensed primary channels in cognitive radio networks. We seek two aims in this paper: (1) estimating the availability probability of each primary channel and (2) admission control of secondary users to decrease the rate of collisions between them. We consider single and multiple secondary user scenarios. In the first scenario, the secondary user deploys learning automata to estimate the primary channel availability probability for efficient exploitation. In the second scenario, each secondary user deploys an algorithm based on MRLA to estimate primary traffic as well as the behavior of other secondary users in order to control the rate of collisions. Then, to have a better control on the rate of secondary collisions, when the number of secondary users is greater than the number of primary channels, we proposed an admission control scheme. In this scheme, some of secondary users are blocked in each time slot and do not have any interaction with the environment. The convergence of the proposed algorithms with and without admission schemes is analyzed. Simulation results are provided to show the improvement in the secondary users' total throughput and switching cost while maintaining the fairness between them.