AI-Enabled Dynamic Spectrum Sharing in Next-Generation Networks

Whereas the current generation networks have been condomized and require efficient spectrum management to support increasing demand for wireless services for next-generation networks (NGNs). This research work suggests the implementation of an AI adapted DSS to manage the utilization of the spectrum efficiently in NGNs. In essence, by applying machine learning, specifically using reinforcement learning, the system adapts to the real-time data and selects the most appropriate resources, thus increasing spectrum efficiency, minimizing interference; and boosting network performance. Network conditions and users’ behaviors of various networks can be obtained based on the plenty of dataset to train the developed AI models. As proved, there have been significant enhancements in LOS rate, spectrum efficiency, throughputs, and latency, showing the effectiveness of AI-based DSS for the future WNs. Several issues in terms of data accessibility, algorithms’ computational cost, security aspect, and the standardization issue are also presented to further stress on the research demands and cooperative work.