Spectrum Sensing is a promising technology that has been proposed as a means of improving the efficiency of wireless communication systems by effectively allocating the radio spectrum among the licensed and unlicensed users. A spectrum sensing scheme uses received signals to detect channel states, and it virtually predicts channel states in the near future simply using previous detected channel states. Intensive work on prediction for cognitive radio has been reported and prediction-based channel sensing also helps to improve the spectrum utilization (SU). Channel status prediction has been carried out in the recent past by employing various prediction models such as MLP neural network, auto regression, etc. For overcoming the disadvantages encountered by previous techniques in channel prediction environment, Correlation and Bayesian probability based LM-neural network (CB-LM) for spectrum sensing in cognitive radio network is proposed in this paper. The proposed technique is implemented using MATLAB and the evaluation parameters employed are SU , imp SU and throughput. The comparative analysis is carried out by comparing to HMM, Bayesian based NN and random technique. From the results, we can see that our technique has achieved greater values for SU , imp SU and throughput in graphs which clearly demonstrates the effectiveness of the technique.
Volume 11 | 08-Special Issue
Pages: 506-524