Decision Fusion Using a Hybrid Spectrum Sensing Technique for Multiple Channels in a Cognitive Radio Network: Evaluation and Simulation

Abstract

Hybrid spectrum sensing techniques that combine matched filtering and cyclostationary feature detection have been shown to improve the detection accuracy and reliability of CRN. MF is a technique that uses a filter coordinated to the shape of the expected signal to perceive the existence of a known signal in noise. It is designed to match the waveform of the primary user signal of interest. CFD is a method for identifying the presence of PU by taking advantage of the regularity in the statistical features of the signal. The CFD looks for the cyclic correlation in the signal, which is the correlation between the signal and a time-delayed version of itself. This combination is particularly useful in environments where the primary user signals are weak or the noise level is high. It provides a robust and reliable way of detecting the occurrence of PU in a dynamic and uncertain environment. The technique can also decrease the number of false detections and get better the overall efficiency of spectrum usage in the CRN.