Optimal Resource Allocation for Energy Harvested Cognitive Radio Networks Based on Learn Heuristic Algorithm

In an Energy Harvested Cognitive Radio Networks (EH-CRN), Primary user (PU) and Secondary User (SU) networks share channel resources. Energy harvesting allows CRN nodes to receive energy from the atmosphere for sustainability. To achieve the best throughput and network capacity. In this paper, author address problems of delayed convergence and the need for huge state spaces in current deep Q-learning-based RA techniques. The RA in EH-CRN is enhanced by the suggested Support Vector Machine based Red Deer algorithm (SVM-RDA), which considers capacity, average latency, and transmission power restrictions. Simulation results suggest the proposed algorithm provides resource utilization and greater convergence than previous methods in the literature.