Real-Time Anomaly Detection in 5G Networks Through Edge Computing

Strong anomaly detection techniques are becoming more and more necessary as 5G networks develop in order to maintain network performance, security, and dependability. This study leverages the capabilities of Mobile Edge Computing (MEC) to present a novel method for anomaly detection in 5G networks. By processing data closer to the network edge, the integration of MEC offers an efficient and decentralized architecture that lowers latency and improves real-time detection capabilities. The distributed module takes advantage of its close proximity to network devices by using sophisticated algorithms for anomaly detection, which are implemented at the mobile edge. The system can quickly detect abnormalities from typical network activity by utilizing capabilities including Flow Collection, Anomaly Symptom Detection, and Network Anomaly Detection. The distributed module provides anomalous information to the centralized decision-making module for thorough examination. It takes into account variables like resource use and network traffic and integrates this data with metrics gathered from monitoring modules. Because of its adaptive characteristics, the system may expand anomaly detection components, enhance detection functions, and modify virtualized resources in response to shifting network circumstances. The evaluation findings reveal that the suggested anomaly detection method performs well in 5G networks, with decreased false positives, increased responsiveness, and better flexibility to changing network conditions. Using MEC not only makes anomaly detection more effective, but it also fits in with the 5G design, which makes it a viable option for protecting the upcoming generation of communication networks. It obtains 95% accuracy in classification. The suggested approach has proven to be resilient in handling security issues by producing outcomes that are either comparable to or better than those attained by other techniques that have been previously presented in the study literature. This demonstrates the model's dependability and effectiveness in handling security-related problems.