Optimal Channel Selection and Switching Using Q-Learning in Cognitive Radio Ad Hoc Networks

With the rising demand for spectrum and the emergence of advanced communication systems, there is a critical requirement for more efficient and streamlined approaches to spectrum utilization. Thus, suitable frequency channel allocation and switching techniques in Cognitive radio (CR) are essential for increasing the spectrum utilization efficiency. Although researchers have been working in this area for a demi-decade, the chances of the collision of primary user and secondary user transmission are still not reduced to zero. To further reduce this problem, the authors in this article have proposed an optimal channel selection and switching strategy for cognitive radio ad hoc networks (CRAHNs) seeking maximum reward for a particular channel using Q-learning algorithm in combination with clustering algorithm. For data transmission, channel with the largest Q-value is chosen. Through extensive simulations and comparative analysis, it can be seen that in comparison to the latest existing scheme, the proposed QLOCA scheme improves packet delivery ratio by 3.8%, throughput is improved by 6.454%, average delay is reduced by 7.2% and packet collision ratio is reduced by 4.2%.