Regression Cooperative Gradient Flying Fox Route Optimization for Energy Efficient Routing in CRN

Cognitive Radio Networks (CRNs) is a formidable evolution in wireless technology owing to its potentiality to resolve spectrum scarcity crisis. In CRNs, cognitive radio user’s sense idle spectrum as well as opportunistically entrée sensed spectrum devoid of approved users. The best spectrum is said to be selected opportunistically, shared with other users with no severity to licensed users. Many CRN devices that are fueled by battery possessing constrained energy capacity. Prevailing battery mechanisms yet couldn’t contribute CRN devices works for a long time. Therefore, conservation of energy is mandatory as far as routing in CRN is concerned. In this work a machine learning based method called, Regression Cooperative Spectrum Sensing and Gradient Flying Fox Route Optimization (RCSS-GFFRO) for energy and bandwidth aware data routing with spectrum utilization in CRNs is proposed. Protection of primary users (PUs) as well as optimization of secondary user (SU) transmission included in proposed method. In the proposed method a cooperative spectrum sensing using machine learning (ML) model called, Median Regression function is applied. The Median Regression Cooperative Spectrum Sensing in our work with the bandwidth and energy efficiency ensures acceptable probability of interference to primary users due to secondary user’s access by measuring the relationship between bandwidth and energy efficiency simultaneously. Next, optimal route identification is performed using Gradient Flying Fox Route Optimization algorithm. By applying gradient optimization function minimizes the interference with primary users and increases secondary user performance via sensing time and transmission time parameters with higher routing performance. Simulation estimated with parameters.