REINFORCEMENT LEARNING APPROACH TO REDUCE LATENCY FOR SPECTRUM SENSING IN COGNITIVE RADIO WIRELESS NETWORKS.

The detection of available wireless channels will allow CR radio transceivers, discovering which communication channels are in use and which are not. The main goal of Cognitive Radio devices is to move into vacant channels while avoiding occupied ones. It passes the transmission of multiple signals into a single medium, optimizing the spectrum while minimizing interference with other users--low latency routing algorithm based on dynamic programming in cognitive wireless mesh networks through modified Q-learning algorithm. This research aims to use an RL technique known as changed Q-Learning to provide a potential solution for allocating channels in a wireless network containing independent cognitive nodes. The proposed method demonstrates the results by spectrum sensing scheme achieves significant performance gain over various reference algorithms in scanning overhead and access delay for particular applications.