Modified Anderson–Darling detector for spectrum sensing

Abstract

Goodness-of-Fit (GoF) testing is a promising approach for blind spectrum sensing in cognitive radio (CR) applications. In this paper, we present a modification on the Anderson-Darling detector by taking into account the physical characteristics of spectrum sensing, resulting in a reduction of the false alarm rate by a factor of two for a given detection threshold in the binary hypothesis test.

Experimental SDR implementation of cooperative spectrum sensing in cognitive radio networks

“…In the literature, several GoF test-based spectrum sensing tests have been proposed. Among the best known are: the Kolmogorov-Smirnov (KS) test [35], the Cramer-Von Mises (CM) test [36], the Anderson-Darling (AD) test [34], Likelihood Ratio (LLR) test [51], and the Modified AD test [37]. All of these experiments are based on the hypothesis test as expressed in (30).…”

“…In this paper, we first implemented four different local spectrum sensing methods for different tests: energy Detection (ED) [12], Maximum-Minimum Eigenvalue (MME) [32], Welch's Periodogram (WP) [33], and a various of Goodness of Fit (GoF) tests based energy detection [34]. In the literature, many GoF tests have been proposed, mainly the Kolmogorov-Smirnov (KS) test [35], Cramer-Von Mises (CM) test [36], Anderson-Darling (AD) test [34], Likelihood Ratio (LLR) test [37], and Modified AD test [37]. The performance of each SS and CSS algorithm was evaluated by depicting the Receiver Operating Characteristic (ROC) (detection probability P d versus false alarm probability P f ) curves, or equivalently, the Complementary ROC (miss detection probability P m = P d − 1, versus P f ), then the feasibility in the real environment was assessed.…”