Fractile-piecewise processing based spectrum sensing algorithms for underwater cognitive acoustics with impulsive noise: [extended abstract]

Abstract

A novel detector is proposed in the framework of fractile-piecewise processing based on the Cauchy-Gaussian (CG) distribution model of impulsive noise. According to the CG distribution, the impulsive noise can be decomposed into two blocks with one representing the Gaussian noise and the other representing the Cauchy noise. The detector processes two blocks separately, where the breakpoint is determined by the sample fractile. Simulations show that the proposed detector outperforms the Fractional Lower Order Moments (FLOM) and Cauchy detector under non-fading channels and Rayleigh fading channels. Besides, the low-computation complexity allows the proposed detector to be an attractive solution in the underwater cognitive acoustic (CA) system.