Sustainable Green Cognitive Radio Networks: Optimized Deep Transfer Learning Model for Energy Consumption

Green sustainability pertains to methods designed to reduce adverse effects on the environment. Aspects of human activity such as energy production, transportation, waste management, agriculture, and industry are all included in the concept of green sustainability. In this context, sustainable cognitive radio networks (CRNs) are designed to optimize both performance and environmental impact. It enables dynamic spectrum access, allowing devices to intelligently utilize available spectrum resources efficiently. The sustainability in CRNs involves reducing energy consumption and maximizing spectrum utilization while preserving reliable communication. This chapter addresses the crucial goal of reducing energy usage in green cognitive radio networks (GCRNs) during communication between secondary users (SUs) and primary users (PUs). The proposed optimized deep transfer learning model in the GCRNs (ODTL-GCRN), based on African Vultures Optimization Algorithm (AVOA), is proposed to enhance spectrum sensing in green CRN based on collaborative Spectrum Sensing. ODTL-GCRN iteratively updates the weight on all layers of the network, and this iterative phase persists until the learned features enable accurate spectrum sensing decisions and reduced energy consumption. ODTL-GCRN leverages knowledge from a specific task to address similar ones. Simulation results highlight ODTL-GCRN effectiveness in reducing energy usage and increasing average effective throughput compared to recent models.