AI-BASED INTRUSION DETECTION SYSTEM FOR EDGE COMPUTING SECURITY

Abstract

Reducing latency and advancing the data processing at the network's edge, one of the technology that has come up to improve the performance of distributed systems is edge computing. However, such a shift poses great security risks that can be addressed only by the use of advanced intrusion detection mechanism. To further improve cybersecurity resilience, this research presents an AI-based Intrusion Detection System (IDS) with regard to edge computing environments. Deep learning algorithms are used in the proposed system for anomaly detection and behavior analysis with better detection accuracy than the traditional methods. The methodology encompasses integrating optimized machine learning models that are trained over the network traffic data with minimal computational overhead, and maximal precision of the detection. Additionally, AI driven chatbots are also included to improve real time alert mechanism and user communication in security events. I demonstrate that the proposed IDS consistently shows performance in detecting known and zero-day attacks and provides these alerts for known attacks using intelligent pattern recognition and adaptive (or machine) learning techniques. Transformative role will be played by recently emerging technologies like Generative Adversarial Networks (GANs) and Graph Neural Networks (GNNs) that assist in ensuring higher robustness of the model against these adversarial attacks. It is found that AI enhanced IDS systems have more accuracy, speed and adaptability compared to traditional rule based methods in dynamic edge computing environments. The contribution of this work is in highlighting that AI is going to be an essential technology to strengthen integrity and security of future edge networks.