Resilience enhancement method of imitative networks based on the artificial immune algorithm

The complexity of data dimensions and network structures has been substantially increased in recent years because of the rapid development of modern networks. This increased complexity, combined with the increasing frequency of internal failures, external attacks, and unexpected events, necessitates the improvement of network resilience and robustness. Mimic defense has emerged as a promising technique for improving network resilience. However, its deployment in large-scale clustered environments encounters challenges such as high overhead, low adaptability, and operational complexity. To address these challenges, in this study, we propose a resilience improvement method for imitative networks based on the artificial immune algorithm (REMI-AIA). The proposed method considers both the defense overhead and load balancing within device clusters while improving adaptive security; it employs a prepost+ algorithm based on long-term spatiotemporal data to correlate attacks with defense mechanisms. AIA iteratively adjusts heterogeneous devices using negative feedback to control their number and generate high-affinity variants in real time, thus forming a dynamic mimic architecture. The simulation results show that, compared with traditional mimic architectures, the proposed REMI-AIA framework reduces defense overhead and considerably improves network resilience and fault tolerance.