Edge Driven Spectrum Optimization in Cognitive Radio Networks for Public Safety and Emergency Communication Systems

Large-scale communication networks like the one in public safety are under the requirements of rapid, reliable, and interference-free spectrum access in the presence of natural disasters, large-scale accidents, and security threats. Olden fixed-spectrum allocation schemes do not fulfill these demands, particularly in cases of emergencies when there is congestion and breakdown of infrastructure systems in the network. The CRNs provide dynamic spectrum access, whereas centralized decision-making provides latency and scale constraints. In this study, a proposed model is an Edge-Driven Spectrum Optimization (EDSO) model which is specific to CRNs in a public safety and emergency communication setting. The advantage of the new model is the use of edge computing nodes to do localized, real-time spectrum sensing, decision making, and allocation to achieve a drastically lower response time and a better spectrum utilization. In contrast to the traditional centralized CRN models, the EDSO model takes advantage of edge intelligence to provide faster response to spectrum availability to enhance resilience during high-stress situations. The proposed EDSO model is evaluated against the conventional centralized CRN models using simulation in MATLAB 2023. The performance metrics analyzed include spectrums sensing accuracy, throughput, utilization efficiency, spectrum, end to end latency and energy consumption. Experiments have shown that EDSO model is better than the current models with up to 35 percent lower latency, 20 percent higher spectrum use, and 25 percent energy consumption in dynamic emergency conditions. This paper is a solid support of how edge intelligence should be incorporated into CRNs to make them more resilient in communication with first responders and the general population safety organizations.