An Autonomic 5G Edge Security Framework via Augmented MAPE-K Loops and Deep learning

Authors:

 Laxmikantha K¹ ,Sanjay T², Sachin R³, Gagan gowda D⁴ , Rakesh R⁵

Laxmikantha K, Assistant Professor, Department of Computer Science & Engineering, K. S. Institute of Technology, Bengaluru

Sanjay T , Department of Computer Science and Engineering, K. S. Institute of Technology, Bengaluru

Sachin R , Department of Computer Science and Engineering, K. S. Institute of Technology, Bengaluru

     Gagan gowda D , Department of Computer Science and Engineering, K. S. Institute of Technology,                                                                                        Bengaluru

     Rakesh R, Department of Computer Science and Engineering, K. S. Institute of Technology, Bengaluru

Abstract - With the 2026 deployment of 5G networks, data transfer speeds reaching 20Gbps have rendered traditional human-centric security systems obsolete due to a critical "Response Gap". This research proposes 5G Block, an autonomic security framework designed for Multi-access Edge Computing (MEC) environments. The system utilizes an augmented MAPE-K loop featuring a high-performance Go-based engine for non-intrusive extraction of 40 flow-based features. Threat detection is achieved through a Long Short-Term Memory (LSTM) neural network trained on the 5G-NIDD dataset, identifying malicious vectors such as DDoS and Port Scans with over 98% accuracy. Uniquely, the framework incorporates an Explainable AI (XAI) layer using SHAP to provide transparency for automated decisions, alongside a private blockchain to anchor decision hashes for immutable forensic auditing. Experimental results demonstrate a detection latency of less than 10ms, successfully meeting the ultra-reliable low-latency requirements of 5G core infrastructure.

Key Words:  5G Security, Blockchain, Deep Learning, MAPE-K, Edge Computing, XAI.