Performance evaluation of neural network based spectrum sensing in cognitive radio

Cognitive radio network has come up as an auspicious solution to spectrum limitation and underutilization over wireless communication challenges. It empowers secondary users to work on licensed spectrum or primary user's spectrum without any interference. Here neural network based spectrum sensing is a significant responsibility for the secondary users to obtain the availability of a vacant channel in the primary user's spectrum. In this paper, main emphasis is to maximize the decision accuracy under different noisy condition and its implementation through the Neural Network system (NNS) that tremendously enhance the desirable throughput. Neural Network is smart intelligent phenomena with the capabilities to understand the environment, learn and adjust in real time operating parameter according to specific need of unlicensed user. Finally, this paper concludes through simulation results after successive approximation, in order to obtain the improved performance with better decision accuracy.