SVM Algorithm for Cooperative Spectrum Sensing in Wireless Fading Channels

The evolution of today’s wireless systems and technologies has enormous impacts on the society. The effect of the developments in wireless communications on the radio frequency spectrum availability demand a considerable amount of attention. The radio frequency spectrum is a finite resource, and its usage should be conducted as efficiently and as maximum as possible. However, the maximum usage of the spectrum is often met with unfavorable conditions, where interference between users of communication channels will be increased. Cognitive radio is a wireless communication system design paradigm that promises to maximize spectrum utilization while decreasing spectrum scarcity. It employs sophisticated software-defined radios and machine learning algorithms to effectively regulate spectrum resources, enabling devices to access underutilized spectrum segments. This research proposes a support vector machine algorithm for cooperative spectrum sensing in wireless fading channels. We employ four secondary users as sensors where different statistical fading values are experienced by different sensors. Performance is evaluated through analysis of accuracy. The simulation yields favorable results based on accuracy values.