This paper surveys the literature relating to the application of machine learning to fault management in cellular networks from an operational perspective. We summarise the main issues as 5G networks evolve, and their implications for fault management. We describe the relevant machine learning techniques through to deep learning, and survey the progress which has been made in their application, based on the building blocks of a typical fault management system. We review recent work to develop the abilities of deep learning systems to explain and justify their recommendations to network operators. We discuss forthcoming changes in network architecture which are likely to impact fault management and offer a vision of how fault management systems can exploit deep learning in the future. We identify a series of research topics for further study in order to achieve this. INDEX TERMS Cellular networks, self healing, cell outage, cell degradation, fault diagnosis, deep learning, explainable AI.

This article presents the results of an investigation of the application of deep learning techniques to the sleeping cell problem, in order to achieve greater detection sensitivity than previously reported. We use a deep recurrent Neural Network (rNN) to process simulated RSRP reports in order to detect degradations of cell radio performance as well as complete outages. Using such a configuration we are able to achieve improved sensivity in relation to a traditional Support Vector Machine (SVM) approach, while eliminating the need for a separate dimensionality reduction stage at the front end. We study multiple rNN configurations with up to three hidden layers and conclude that in this scenario we can achieve the target improvement in sensitivity with a single hidden layer, leading to highly efficient run time performance.

No abstract