Artificial Intelligence Enabled Self-healing for Mobile Network Automation

This paper presents an artificial intelligence enabled self-healing framework for cell outage detection and compensation in radio access networks. The developed framework consists of three modules, namely cell outage detection, cell outage compensation, and continuous optimization that work in closed-loop to detect outages, trigger recovery actions, and network optimization to minimize the impact of outages on user experience. The outage detection module is based on machine learning algorithms aimed to detect anomalies in the network performance data. Likewise, the cell outage compensation module uses fuzzy logic to determine compensation actions after an outage cell has been detected. The continuous optimization module is tasked with making incremental improvements to the network configuration through a heuristic approach. The developed self-healing framework is validated using a network simulator ns-3 based test environment. Results show the framework is capable of fully recovering from the outage in terms of accessibility and coverage. In addition, the cell edge reference signal received power is recovered by 45%, thereby significantly improving the network performance once the outage is detected.