AdaptiveMECShield: A trust-based, threshold-aware framework for real-time DDoS mitigation in mobile edge computing

The advancement of Mobile Edge Computing (MEC) technology has improved the effectiveness of edge networks by offering lower latency and increased computational efficiency. Nonetheless, MEC technology continues to be susceptible to Distributed Denial of Service (DDoS) attacks, which degrades it's performance, overwhelm resources, and threaten network availability. This study presents AdaptiveMECShield, an innovative DDoS mitigation scheme designed to improve the resilience of MEC against DDoS attacks. AdaptiveMECShield utilizes a multi-tiered mitigation approach that combines adaptive load balancing, detection based on dynamic thresholds, and real-time traffic surveillance to effectively detect and address anomalies. The suggested scheme utilizes a windowing method for detecting anomalies and incorporates lightweight controllers to facilitate effective decision-making without degrading MEC resources like CPU and memory. The performance of AdaptiveMECShield is evaluated through comprehensive simulations in regulated MEC settings. The findings indicate that the scheme achieves a 92.5 % detection accuracy, keeps the false positive rate at 3 %, and reduces false negatives to 2 %, while allowing early attack detection in under 10 simulation seconds. Furthermore, the proposed method shows effective CPU and memory usage, guaranteeing that MEC servers are not overburdened during DDoS mitigation. The findings show that AdaptiveMECShield offers a proactive, adaptive, and resource- efficient approach in countering DDoS attacks in real time edge computing environments. This research contributes to the design of strong security measures for MEC, improving network durability and guaranteeing consistent service provision despite changing cyber attacks. Future efforts may focus on deploying AdaptiveMECShield in real workd MEC and investigating more methods to further enhance resource utilization and improve scalability.