The dynamic architecture of 5G networks, defined by Software-Defined Networking (SDN) an‎d Network Functions Virtualization (NFV), offers unparalleled flexibility while also presenting intricate operational issues an‎d susceptibility to anomalies. Conventional fault management techniques are inadequate for real-time detection an‎d recovery in this context. This concept introduces an AI-driven self-healing architecture that independently detects, diagnoses, an‎d resolves problems in 5G software-defined mobile networks. The proposed system utilizes both supervised an‎d unsupervised machine learning models, such as Ran‎dom Forest, LSTM, CNN, an‎d Autoencoders, in conjunction with SDN controllers an‎d NFV orchestrators. The system utilizes feedback loops an‎d reinforcement learning to progressively enhance rehabilitation procedures. eva‎luation measures encompass detection latency, classification accuracy, an‎d the stability of quality of service (QoS) in fault scenarios. The expected result is a robust, low-latency, autonomous self-healing network architecture that adheres to Zero-Touch Service Management (ZSM) principles an‎d facilitates future 6G advancements.

6/10/2025 12:00:00 AM

86986

1404/04/31

AI-Driven Self-Healing Mechanisms for Real-Time Anomaly Detection in 5G Software-Defined Mobile Networks

5G networks , Software-Defined Networking (SDN) , Network Functions Virtualization (NFV), Anomaly detection , Self-healing systems, Artificial Intelligence (AI) , Zero-Touch Network Management