Akademik Veri Yönetim Sistemi
26th IEEE Signal Processing and Communications Applications Conference (SIU), İzmir, Türkiye, 2 - 05 Mayıs 2018
In this study, for channels containing primary user traffic, we investigate how non-Gaussian noise model affects detection performance in cognitive radio. As decision statistic, energy detector has been preferred by considering its simple structure and the fact that it does not need a priori knowledge about primary user signal. Gaussian mixture model, Laplacian distribution, and symmetric alpha-stable distribution have been used as non-Gaussian noise models. The status changes of the primary user have been modeled via Poisson process. The simulations have been carried out separately for both under status changes and no status changes in the channel. According to our simulation results, primary user traffic degrades detection performance as does existence of non-Gaussian noise. In addition, it has been observed that the maximum degradation in detection performance of energy detector occurs for symmetric alpha-stable distributions among the considered noise models in this study.