A learning-based approach for autonomous outage detection and coverage optimization

Abstract

To be able to provide uninterrupted high quality of experience to the subscribers, operators must ensure high reliability of their networks while aiming for zero downtime. With the growing complexity of the networks, there exists unprecedented challenges in network optimization and planning, especially activities such as cell outage detection (COD) and mitigation that are labour-intensive and costly. In this paper, we address the challenge of autonomous COD and cell outage compensation in self-organising networks (SON). COD is a pre-requisite to trigger fully automated self-healing recovery actions following cell outages or network failures. A special case of cell outage, referred to as sleeping cell, remains particularly challenging to detect in state-of-the-art SON, because it triggers no alarms for operation and maintenance entity. Consequently, no SON compensation function can be launched unless site visits or drive tests are performed, or complaints are received by affected customers. To address this issue, our COD solution leverages minimization of drive test functionality, recently specified in third generation partnership project Release 10 for LTE networks, in conjunction with state-of-the art machine learning methods. Subsequently, the proposed cell outage compensation mechanism utilises fuzzy-based reinforcement learning mechanism to fill the coverage gap and improve the quality of service, for the users in the identified outage zone, by reconfiguring the antenna and power parameters of the neighbouring cells. The simulation results show that the proposed framework can detect cell outage situations in an autonomous fashion and also compensate for the detected outage in a reliable manner.