Dynamic Robust Spectrum Sensing Based on Goodness-of-Fit Test Using Bilateral Hypotheses

Abstract

Dynamic spectrum detection has attracted increasing interest in drone or drone controller detection problems. Spectrum sensing as a promising solution allows us to provide a dynamic spectrum map within the target frequency band by estimating the occupied sub-bands in a specific period. In this paper, a robust Student’s t-distribution model is built to tackle the scenario with a small number of observed samples. Then, relying on the characteristics of the statistical model, we propose an appropriate goodness-of-fit (GoF) test statistic regarding a small number of samples. Moreover, to obtain a reliable sensing, bilateral hypotheses of the test statistic are both used to make a decision. Numerical simulations show the superiority of the proposed method compared with other schemes, including the unilateral hypothesis-based GoF testing and the conventional energy detection, in a small number of sample cases.

Performance Analysis of Centralized Cooperative Schemes for Compressed Sensing

“…This section considers the SR case, where the SUs do not make any hard decision and transmit soft information to the FC, specifically the energy values collected by the EDs. According to (10), the FC combines the received energy values using EGC, therefore the FC acts like a single ED with MS samples, where S is the number of active SUs and M is the compressed sample size of each ED.…”

“…A second major factor of CSS is the sensing algorithm of the SUs. The choice of this algorithm mostly depends on the available information about the structure of the PU signal [ 6 , 7 , 8 , 9 , 10 , 11 , 12 , 13 ]. Matched filtering and feature extraction presuppose that the SUs have some knowledge about the PU signal properties.…”

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

“…Researchers have suggested that analyzing the GoF may be a promising new approach to detecting issues, especially given the limitations of even the most advanced detection methods currently available. The authors have suggested numerous GoF tests in mathematical statistics literature, such as the Kolmogorov-Smirnov(KS), Cramer-Von Mises(CM), Shapiro-Wilk, and Anderson Darling(AD) tests [16] [17], which are used to quantify the dissimilarity between two distribution functions during the presence and absence of a signal. In the study described in [18], the KS test, a non-parametric method for GoF analysis, was employed to rapidly and reliably sense spectral data.…”