Anomaly Detection in Self-Organizing Mobile Networks Motivated by Quality of Experience

In today's self-organizing mobile communication networks, the approaches that are used for anomaly detection use network-centric tactics to find faulty service nodes. These methods were previously used. These methods are now being used in the process of anomaly detection. This investigation presents a user-centric strategy and a unique technique for identifying anomalies as potential solutions. Both of these approaches are innovative. The Quality of Experience (QoE) metric is used to determine how the end user should rate a product or service's overall performance product or service's overall performance should be rated by the end user. The abbreviation for “quality of experience” (QoE) is “QoE.” Using a descriptive Quality of Experience model and machine learning to predict user Quality of Experience in a network scenario constructed using the ns-3 simulation tool, the system model effectively illustrates how malfunctioning serving eNodeBs may be detected. The purpose of this was to show that faulty serving eNodeBs may be identified. An experiment like this was conducted to demonstrate how inefficient serving eNodeBs may be found. Future ultra-dense and eco-friendly mobile communication networks may benefit greatly from this strategy. It is also expected that these networks can self-organize and repair themselves if anything goes wrong.