Convolution Based Energy Detection Scheme for Cognitive Radio Systems

Abstract

In this paper, the authors investigate an energy detection scheme for Cognitive Radio (CR). The mathematical analysis of the probability density function (pdf) based Likelihood Ratio Test (LRT) statistic for deciding in favor of Primary User (PU) absence or presence proceeds by means of convolution statistics. Dynamic threshold is set for comparing with LRT test based on fixed probability of false alarm and different number of samples. Compressive Sensing (CS) based minimum required number of samples and Central Limit Theorem (CLT) based equivalent number are the two comparison benchmarks, based on Gaussian statistics. The maximum percentage from this cases is derived, in order to compare with the maximum percentage of samples above threshold to the total number of samples for the Rayleigh, Rician and Nakagami-m fading channels. A related algorithm is provided along with technical interpretation of simulation results. Conclusions finalize the paper.