Machine Learning Aided Context-Aware Self-Healing Management for Ultra Dense Networks With QoS Provisions

Abstract

The self-organizing network is envisioned as a key technology to future wireless networks, especially for densely deployed small cell scenarios. Self-healing (SH) is an essential functionality to allow the networks to automatically detect and compensate for cell outages, which typically occur when unexpected network failures arise. In this paper, reaping the benefits of machine learning, we propose a novel SH framework in ultra dense small cell networks for meeting the demands of low-cost and fast network operation, quality of service (QoS), and energy efficiency. The proposed SH scheme comprises small cell outage detection (SCOD) and small cell outage compensation (SCOC) to enable self-healing in ultra dense small cell networks. Based on the context information of the partial key performance indicator (KPI) statistics, we propose a novel SCOD algorithm to detect the outage by applying support vector data description (SVDD) approach. The SCOD algorithm detects a small cell outage efficiently considering two situations: KPIs available situation and non-KPIs available situation. Furthermore, in order to compensate the small cell outage, SCOC is formulated as a network utility maximization problem to optimally compensate for the outaged zone in small cell network. A distributed compensation algorithm with low computational complexity is developed to balance the load of small cell networks, considering the QoS provision for users. Simulation results demonstrate that the proposed SH scheme can detect the small cell outage efficiently and can achieve an optimized QoS performance when compensating for the detected small cell outage.