Vulnerability Analysis for Network Connectivity: A Prioritizing Critical Area Approach

Analyzing network vulnerability, especially connectivity vulnerability, is vital for network security planning. Traditionally, network vulnerability analysis methods separate the studies of global connectivity vulnerability and critical area vulnerability, and thus ignore joint failure of network connectivity and critical-area integrity that may cause grave damage to a network. To this end, this paper proposes a prioritizing critical area approach for connectivity analysis to identify the corresponding vulnerable elements. Specifically, we consider the worst-case scenario of a network and aim at finding the minimum disruption-cost set of elements whose removal not only severely damages network connectivity but also disrupts the critical-area integrity. Since the above optimization problem is NP-hard, a heuristic algorithm based on spectral partitioning is developed to solve it. Simulation results validate the effectiveness of our proposed scheme in accurately identifying the vulnerable elements in critical areas to prevent significant loss in the overall network connectivity and performance.