A Cyclostationarity-Based Spectrum Sensing Scheme for Cognitive Radios With Dynamic Primary User Signals

This paper addresses a cyclostationarity-based spectrum sensing scheme for cognitive radios in dynamic primary user (PU) traffic environments where the PU might randomly depart or arrive during the sensing period. At first, the spectrum sensing problem in dynamic PU traffic environments is formulated as a binary hypothesis testing problem. Then, a test statistic for spectrum sensing is derived by applying an estimate of spectral coherence function of the PU signal to the generalized likelihood ratio. Numerical results show that the proposed scheme exhibits a better spectrum sensing performance than that of the conventional scheme based on the energy detection in the presence of dynamic PU signals.