Artificial Neural Network Based Approach for Spectrum Sensing in Cognitive Radio

In the recent wireless scenario, Cognitive Radio (CR) is a critical technique that supports efficient utilization of the radio spectrum. Spectrum sensing is the main task of CR that plays a main role in deciding the spectrum availability. In this paper, Artificial Neural Network (ANN) based spectrum sensing (SS) in Cognitive Radio is proposed. Various spectral features of received signals are measured to create a database to train the ANN. The trained ANN is then used to classify the signal and noise samples. Simulation results obtained show that the proposed technique detects the signal under considerably poor Signal to Noise Ratio (SNR) scenario. ANN based spectrum sensing exhibits reliable performance compared to conventional energy detection based spectrum sensing.