A Novel Spectrum Sensing Method with Generative Adversarial Networks in Cognitive Radio Network for UAV

Abstract

In cognitive radio (CR) networks, spectrum sensing is an important fundament function and energy detection is a popular technique in spectrum sensing to detect primary user (PU). Since its low computational and implementation complexities, energy detection is the most common used technique for signal detection. This paper studies detection performance of primary signal under additive white Gaussian noise (AWGN) to propose a novel spectrum sensing method with generative adversarial networks. The performance of this proposed method is evaluated through a curve with detection probability and simulation time. Then the detection probability in different SNR is also studied. Also, we compared out proposed method with an optimal energy detection method. The derived results show that the proposed method works better than the optimal energy detection method in low SNR situation and as well as in a scenario of mobility.