Modified Bayesian algorithm‐based compressive sampling for wideband spectrum sensing in cognitive radio network using wavelet transform

Abstract

SummaryThis paper presents the implementation of a modified version of Bayesian relevance vector machine (RVM)‐based compressive sensing method on cognitive radio network with wavelet transform for spectrum hole detection. Bayesian compressive sensing is used in this work to deal with the complexity and uncertainty of the process. The dependency of the Bayesian compressive sensing on the knowledge of noise levels in the measurement has been relaxed through the proposed Bayesian RVM‐based compressive sensing algorithm. This technique recovers the wideband signals even with fewer measurements maintaining considerably good accuracy and speed. Wavelet transform is used in this paper to enable the detection of primary user (PU) even in the low regulated transmission from unlicensed user. The advantage of this approach lies in the fact that it enables the evaluation of all possible hypotheses simultaneously in the global optimization framework. Simulation study is performed to evaluate the efficacy of the proposed technique over the cognitive radio environment. The performance of the proposed technique is compared with the conventional Bayesian approach on the basis of recovery error, recovery time and covariance to verify its superiority.