Effect of Non-Gaussian Noise on Detection Performance of Multi-Stage Detectors

Abstract

It is well known that detection performances of conventional detection methods deteriorate under non-Gauss noise. This problem is also encountered in cognitive radio where impulsive nature of noise negatively affects spectrum sensing. Multistage detection techniques can be considered as a possible solution as they analyse signal samples using different detection techniques in a sequential manner. In this study, spectrum sensing performance of a multistage detection method is investigated for detection of a dynamic primary user signal contaminated by non-Gaussian noise. At first, observed signal samples are analysed by energy detector and an ensuing absolute value cumulation computation. Then, a decision as to whether the channel is in use or not is given based on the result of this analysis. Accordingly, if any one of those techniques detects the primary user signal, then the channel is assigned as busy.