Network Survivability Evaluation Model Based on Immune Evolution and Multiple Criteria Decision Making

Abstract

Network survivability has the characteristics of complexity, dynamic evolution and uncertainty, which has become one of the most important indicators of evaluating network performance. Network survivability evaluation is a process of confirming the degree to which the network survivability technology and mechanism can defend network threats. Firstly, it needs to extract useful information from the actual system and establish the model, then the model is analyzed and necessary information is extracted. Finally, the survivability characteristics of the network can be measured and evaluated based on the model and extracted information. Current network security research lacks of models and methods which guide network survivability evaluation effectively. In this paper, we firstly describe the elements of network survivability evaluation model and their relationships, then propose a network survivability evaluation model based on evolutionary computation which includes the immune evolutionary algorithm for network survivability metric weight solving and network survivability evaluation method based on multiple criteria decision making. Finally, the correctness and effectiveness of the proposed model is validated through experimental analysis.