Brain-Inspired Real Time Anomaly Detection System for Mobile Networks

The rapid increase in mobile data traffic, coupled with the growing number of connected devices and applications in networks, puts a substantial strain on existing network management approaches that are highly dependent on human operators. Consequently, there is a pressing need for an automated network management system capable of effectively predicting and detecting anomalies. Presently, the most pow-erful methods used to perform anomaly detection in mobile networks are based on deep learning. However, deep learning-based approaches are not without drawbacks. Specifically, deep learning faces challenges related to noise, offline training, and the volume of training data needed. In this paper, we investigate the potential of the Hierarchical Temporal Memory (HTM) algorithm for performing anomaly detection in mobile networks. This choice is motivated by HTM's advantageous properties, including noise tolerance and online learning capabilities. We conduct a thorough performance evaluation of the proposed anomaly detection system based on HTM. Our model had an F1 score exceeding 0.8, demonstrating a substantial improvement over the current state-of-the-art methods.