Designing a secure access network technique to mitigate the impact of distributed denial of service attacks on mobile edge computing

Abstract

Mobile Edge Computing (MEC) enhances networking technology by bringing computing and storage close to end users. However, this advancement introduces security challenges, particularly its susceptibility to Distributed Denial of Service (DDoS) attacks that threaten network availability and performance in MEC. This study tackles these security vulnerabilities by introducing AdaptiveMECShield, an anomaly-based detection system designed to mitigate DDoS attacks in MEC environments. AdaptiveMECShield integrates several key features to efficiently detect and respond to malicious activity, load balancing, adaptive thresholding, windowing strategies, and real-time traffic monitoring using Software Defined Networking (SDN) controllers. The system dynamically adjusts trust values for network nodes based on packet header information, enabling timely and accurate detection of abnormal traffic. The adaptive thresholding mechanism allows the system to effectively respond to varying traffic patterns, minimizing false positives and negatives while maintaining high detection accuracy. Rigorous simulations demonstrated that AdaptiveMECShield outperformed the Self-Organising Map (SOM) scheme in terms of detection accuracy, false positive and negative rates, detection times, and resource consumption efficiency. In conclusion, AdaptiveMECShield significantly improves the security of MEC systems against DDoS attacks. Its proactive and adaptive nature ensures enhanced security and resilience of edge computing and IoT networks, contributing to uninterrupted availability and improved performance of critical services.