Rapid cascading risk assessment and vulnerable satellite identification schemes for LEO satellite networks

LEO satellite networks are increasingly recognized as a vital component of critical infrastructure worldwide. However, their vulnerability to cascading failures, triggered by both natural and human factors, remains a significant concern. In these networks, failures can propagate rapidly, highlighting the need for quick and accurate assessment of cascading risks and the protection of critical satellites to enhance network reliability. Despite this, current research predominantly focuses on analyzing complex cascading patterns using high-complexity algorithms, which are often not timely enough for effective risk assessment. To address this issue, this paper proposes a rapid cascading risk assessment scheme. It introduces a satellite dynamic simplification method that combines physical and logical addressing and provides an abstract model for assessing failure propagation risk, alongside a quadratic-order algorithm. Furthermore, a scheme for identifying vulnerable satellites is proposed, including the establishment of an optimization model, analysis of the applicability of a greedy strategy, and the development of a vulnerable satellite identification algorithm based on this strategy, achieving cubic-order complexity. Extensive case studies conducted on three constellations of different scales demonstrate that our risk assessment scheme can achieve over 90% accuracy with a deviation of less than 6.2% in 2 s.The vulnerable satellite identification scheme accurately identifies satellites likely to cause severe cascading failures within 7.5 s. The proposed schemes can be implemented in satellites or ground stations to provide rapid risk decision support during unexpected events, offering practical tools for network managers to enhance network reliability.