An Online Game-Integrated Auction Scheme for DDoS Mitigation in Mobile Edge Computing

With the wide implementation of mobile edge computing (MEC) in latency-sensitive applications, edge distributed denial-of-service (DDoS) attacks have emerged as a critical threat to system availability. Existing edge DDoS mitigation approaches predominantly rely on quasi-static configurations, making them inadequate in handling the dynamic evolution of user requests, resource availability, and attack behaviors over time. Ignoring system dynamics can lead to strategy inefficiency, resource wastage, and service degradation. To address these challenges, we propose the dynamic edge DDoS mitigation (DEDM) problem and design an online DEDM framework, OLEDM. It models DEDM as a multi-slot dynamic decision-making process, builds a potential game for request competition in each time slot, and achieves stable request evacuation through multi-player best response updates. Across time slots, it introduces an auction-based strategy adoption mechanism to dynamically decide whether to adopt new strategies, thereby adapting to continuous changes in request patterns and resource conditions. Extensive experiments and theoretical analysis demonstrate that OLEDM outperforms existing methods by effectively adapting to dynamic variations in request load, attack intensity, and edge resource availability.