A New Spectrum Sensing Method with Low SNR under Laplace Noise

Abstract

Aiming at the existing spectrum detection technology, under the condition of low SNR, especially under the background of non-Gaussian noise, the detection performance is low and the detection time is long. This paper fully combines the advantages of energy detection and sequential detection. According to the characteristics of non-Gaussian noise, a new method of spectrum sensing with low SNR under Laplace noise is proposed. For the proposed method, we first divide the received signal equidistantly in the background of Laplace noise, and then use the sequential detection method to achieve a predetermined good detection probability. The simulation results verify the correctness of our theoretical derivation and show that the proposed scheme is correct. Compared with energy detection under Laplacian noise, this method has the advantages of short detection time, small sample size, small influence of noise uncertainty, and good performance under low SNR. In addition, the proposed method can effectively counter the influence of fading channels on detection performance.