Adaptive Sensing Policies for Cognitive Wireless Networks using Learning Automata

This paper introduces an adaptive spectrum sensing method for cognitive radio wireless networks. The proposed method enhances previously proposed random-based sensing policies, effectively selecting the licensed channels to be sensed by accurately estimating channels' availability, resulting, thus, to high system's resources utilization. The core mechanism of the adaptive method is an enhanced learning automaton, which efficiently interacts with the environment and provides accurate decisions on selecting the channel to be sensed on behalf of the secondary users. A thorough description of the introduced method is provided, while the performance of the enhanced sensing policies is verified through extensive simulation experiment.