Self-Healing and Resilience in Future 5G Cognitive Autonomous Networks

In the Self-Organizing Networks (SON) concept, self-healing functions are used to detect, diagnose and recover degraded states in the managed Network Functions (NFs) or other resources. Such methods are increasingly important in future network deployments, since ultra-high reliability is one of the key requirements for the future 5G networks, for example in critical machine-type communication and in other use cases. In this paper, we discuss the considerations for improving the resiliency of future cognitive autonomous mobile networks. In particular, we present an automated anomaly detection and diagnosis function for SON self-healing based on statistical methods, Case-Based Reasoning (CBR) and a concept called augmented diagnosis, which is an active learning method mitigating the collection and maintenance effort of a CBR diagnosis knowledgebase and optimizing its quality. In augmented diagnosis, insights from both the human expert and sophisticated unsupervised and semi-supervised machine learning methods are collected and combined in an iterative way. Additionally, we discuss a more holistic view on mobile network self-healing and utilizing transfer learning methods in the diagnosis.