INTELLIGENT DIAGNOSIS AND OPTIMIZATION OF WIRELESS NETWORKS USING MACHINE LEARNING

The rapid evolution of wireless networks, particularly with the advent of 5G and the emerging 6G paradigm, has introduced significant challenges in network management, fault diagnosis, and performance optimization.Traditional rule-based network management systems struggle to adapt to dynamic and heterogeneous environments. This paper proposes an intelligent machine learning (ML)-driven framework for automated diagnosis and optimization of wireless networks. The proposed approach combines supervised learning techniques for fault classification and unsupervised learning for anomaly detection with reinforcement learning based optimization strategies to enhance Quality of Service (QoS) parameters. Experimental evaluation on simulated datasets demonstrates improved diagnosis accuracy, increased through put,and reduced latency compared to traditional approaches, highlighting the potential of ML in next-generation wireless network management.