Detection of Degraded Wi-Fi APs Using ML-Based Models and Recovery Strategies

With the increasing demand for stable and highperformance wireless networks, ensuring the reliability of Wi-Fi Access Points (APs) has become a critical challenge. Traditional network management relies heavily on reactive troubleshooting, which can lead to prolonged service disruptions and increased maintenance costs. This paper proposes a Machine Learning (ML)-based diagnostic and recovery mechanism for AP degradation, leveraging periodically reported data from APs. By analyzing key performance indicators, the model predicts potential AP degradation, enabling the system to take appropriate actions, such as remote recovery or hardware replacement. The proposed approach improves network resilience by enabling proactive maintenance, reducing downtime, and optimizing operational efficiency. Experimental results demonstrate the effectiveness of the ML-based model in accurately diagnosing AP degradation and enhancing network reliability. This study contributes to the advancement of intelligent network management by integrating predictive analytics with automated recovery mechanisms.