Machine Learning-Based Cooperative Spectrum Sensing in A Generalized α-κ-μ Fading Channel

Abstract

An improvement in spectrum usage is possible with the help of a cognitive radio network, which allows secondary users’ access to the unused licensed frequency band of a primary user. Thus, spectrum sensing is a fundamental concept in cognitive radio networks. In recent years, Cooperative spectrum sensing using machine learning has garnered a great deal of attention as a technique of enhancing sensing capability. In this study, K-means clustering is taken into consideration for the purpose of analyzing the effectiveness of cooperative spectrum sensing in a generalized α-κ-μ fading channel. The proposed approach is examined using receiver operating characteristic curves to determine its performance. The effectiveness of the proposed strategy is contrasted with that of the existing detection techniques such as Cooperating spectrum sensing based on energy detection and OR-fusion-based cooperative spectrum sensing for fading channels κ-μ, α-κ-μ. As demonstrated by results, the proposed method outshines an existing method in terms of comparison parameters, as determined by simulation results in the MATLAB version.