Performance comparison of energy, matched-filter and cyclostationarity-based spectrum sensing

This paper presents a comprehensive performance comparison of energy detection, matched-filter detection, and cyclostationarity-based detection, the three popular choices for spectrum sensing by cognitive radios. Analytical expressions for the false alarm and detection probability achieved by all the detectors are derived. For cyclostationarity-based detection, two architectures that exploit cyclostationarity are proposed: the Spectral Correlation Density (SCD) detector, and the Magnitude Squared Coherence (MSC) detector. The MSC detector offers improved performance compared to existing detectors, and this is demonstrated using the 802.22 RF capture database. It is also shown that the cyclostationarity-based detectors are naturally insensitive to uncertainty in the noise variance, as the decision statistic is based on the noise rejection property of the cyclostationary spectrum. Simulation results plotting the receiver operating characteristics corroborate the theoretical results, and enable visual comparison of the performance.