Neuromorphic AI Empowered Root Cause Analysis of Faults in Emerging Networks
Mobile cellular network operators spend nearly a quarter of their revenue on network maintenance and management. A significant portion of that budget is spent on resolving faults diagnosed in the system that disrupt or degrade cellular services. Historically, the operations to detect, diagnose and resolve issues were carried out by human experts. However, with diversifying cell types, increased complexity and growing cell density, this methodology is becoming less viable, both technically and financially. To cope with this problem, in recent years, research on self-healing solutions has gained significant momentum. One of the most desirable features of the self-healing paradigm is automated fault diagnosis. While several fault detection and diagnosis machine learning models have been proposed recently, these schemes have one common tenancy of relying on human expert contribution for fault diagnosis and prediction in one way or another. In this paper, we propose an AI-based fault diagnosis solution that offers a key step towards a completely automated self-healing system without requiring human expert input. The proposed solution leverages Random Forests classifier, Convolutional Neural Network and neuromorphic based deep learning model which uses RSRP map images of faults generated. We compare the performance of the proposed solution against state-of-the-art solution in literature that mostly use Naive Bayes models, while considering seven different fault types. Results show that neuromorphic computing model achieves high classification accuracy as compared to the other models even with relatively small training data