Dependence analysis based cyber-physical security assessment for critical infrastructure networks
Xiaoxue Liu, Jiexin Zhang, Peidong Zhu · 2016
Critical infrastructures like smart grid and smart transportations are large-scale distributed CPSs and cyber-physical security is pivotal to their safety and security. Efficient security assessment can help administrators to protect CINs effectively and we propose a universal quantitative cyber-physical security assessment methodology which is based on cyber-physical dependence analyses. For given CPS, we first construct the cyber-physical dependence model and dependence probability matrixes to analyze the dependences among assets quantitatively. Then we introduce security indexes and define the cyber-physical security metric ACPI to quantify the cyber-physical security state of each node in the CPS. Furthermore, we design the cyber-physical security assessment algorithm which can compute the ACPI values and discovery potential threats propagating paths in the CPS. Our methodology can be used for distributed CPS by dividing the whole CPS into smaller sub-ones and applying the same methods and rules. Finally, simulations demonstrate the high time efficiency of our assessment algorithm.