Analysis of Wideband spectrum sensing using Anderson darling test statistics in underlay cognitive microcell

Abstract

The most widely used sensing technique for detecting energy is due to its ease of use and lack of need for signal information beforehand. Traditional methods often employ a two-threshold strategy in low signal-to-noise ratio scenarios, which can lead to temporary detection outcomes when the signal's energy is between the low and high threshold values. This subsequently results in inadequate detection accuracy, characterized by a reduced probability of detection, which ultimately results in an extended duration of the spectrum sensing. The paper suggests enhancing the efficiency of spectrum sensing through the utilization of a modified version of the Anderson-Darling test statistic, as opposed to other test statistics.To detect the presence of licensed users in each subchannel from the base station, the cognitive user implements the Anderson Darling test statistics as a sensing technique. The proposed work involves the mathematical derivation of the Anderson-Darling test statistics under a bandwidth of 6MHz.The simulation outcomes show that the sensing technique presented in this research study achieved a detection probability of 0.963 at a false alarm probability of 1%, using a small sample size of 20 at a signal-to-noise ratio of -20dB. This performance outperformed the other three detection algorithms.