Combining Fuzzy Logic and Reinforcement Learning for Resource Management in Edge Computing

With the wide spread of Internet of Things (loT) devices, edge computing plays a decisive part in meeting the computational demands of loT applications. Efficient resource allocation in such networks is essential to ensure optimal utilization of resources and meet the quality of service (QoS) necessities of loT applications. This research addresses the challenge of dynamic resource allocation in edge computing networks for loT applications through a novel hybrid AI approach, “FledegeRL.” By integrating fuzzy logic and reinforcement learning, FledegeRL aims to optimize resource utilization and enhance system performance. The proposed framework is evaluated through simulation analysis and compared with existing algorithms, including Q-Iearning, Fuzzy Logic Control, and Ant Colony Optimization, using metrics such as resource allocation accuracy, average throughput, and network overhead. The results demonstrate that FledegeRL consistently outperforms these algorithms, achieving higher accuracy, greater throughput, and lower network overhead. This highlights the potential of FledegeRL as an effective solution for dynamic resource allocation in edge computing environments.