Optimization of Spectrum Handoff in Cognitive Radio Networks using Markov Decision Processes and Deep Learning

Abstract

The goal of dynamic spectrum access is to resolve the recent concern of remote spectrum inefficiency while also meeting the expanding demands of wireless networks. As a result, a new field of study and development is created for Cognitive Radio (CR) technology, which is essential for utilizing underutilized spectrum through dynamic spectrum access. Therefore, it is guessed that future vehicular correspondence would be CR empowered, utilizing more spectrum options to increase the effectiveness of vehicular communication. Spectrum handoff is a way to dynamically use underused spectrum. A few radio access network could exist together in the execution of CR vehicular network later on. It's conceivable that these network vary essentially in various ways. Consequently, picking the best organization for the spectrum handoff decision among a few radio access networks with differing qualities as far as different boundaries turns into a difficult undertaking for the CR vehicular hub. This supports the utilization of Multiple Attributes Decision Making (MADM) methods to think up a spectrum handoff technique for the best organization choice in CR vehicular network. The spectrum handoff procedure for the best organization determination in CR vehicular networkis the subject of this paper. Given the preferences of CR vehicular nodes, the system offers a greater and optimal selection among the various networks that are available.