Performance Validation of Spectrum Sensing Using Kernelized Support Vector Machine Transformation

The constant development of interest experienced by wireless networks makes a spectrum accessibility challenge. Cognitive radio (CR) is a promising solution to overcome the challenges in spectrum utilization. The process of finding the spectrum holes (availability of spectrum) is called spectrum sensing (SS) which is a major task in cognitive radio network (CRN). Various creative methodologies are proposed in the literature to track the spectrum and identify the available holes. The use of Machine Learning strategies for spectrum sensing has attracted interest in the literature. Therefore, we have considered an ML technique, Support vector machine (SVM) method with Kernel transformation to achieve better results in spectrum sensing. The proposed conspire astoundingly further develops the spectrum detecting execution, but also essentially builds the open doors for dynamic access to the licensed spectrum for the unlicensed users.