Scheduling algorithm based on heterogeneity and confidence for mimic defense

Abstract

As a defense technology with endogenous security, mimic defense plays an important role in network security research. The scheduling of executors is one of the severe problems to take into account for mimic defense, and current research lacks comprehensive consideration of the influence of system architecture and attack behavior on scheduling algorithm. Based on previous research, this paper first introduces concept of heterogeneity and confidence according to vulnerability attributes and attack distribution characteristics to characterize the executors. Moreover, the TOPSIS (Technique for Order Preference by Similarity to an Ideal Solution) algorithm is brought in to optimize the system security and improve operating efficiency. Experimental results showed that,compared with the existing algorithms, Random, MD, RSMS, it improves the security of the system in non-uniform distributed attack scenario and the operating efficiency in each attack scenario.