A Network-Level 3-Layer Single Event Upset Risk Assessment Model for Satellite Mega Constellations

Satellite Mega Constellations (SMCs) are emerging as a critical infrastructure for future space-based 6 G networks, remote sensing, and global navigation systems. With the increasing scale of satellite constellations, SMCs operating in complex and dynamic space environments are becoming increasingly vulnerable to Single Event Upset (SEU). However, the damage mechanisms of SEU within SMCs remain unclear, making it difficult to accurately evaluate their network-level service capabilities. Therefore, this paper aims to establish a risk assessment model that captures the evolution of SEU damage from component, to system, and eventually to the network level for SMCs. This paper proposes a network-level 3-Layer SEU risk assessment model for SMCs, which begins by modeling component-level SEU risk at various spatial locations based on the Poisson characteristics of SEU occurrences. Next, a continuous time Markov chain (CTMC) is employed to model the interactions between components and the evolution of the satellite communication system states, deriving the contribution of each state to the system-level SEU risk. Finally, by integrating the hazard (H), severity (S), and vulnerability (V) required by the HSV model using an information entropy function, risk values for each node in the SMCs are assessed, thereby enabling the evaluation of SEU risks in SMCs at the network level. Building on this, this paper investigates the relationship between altitude, inclination, constellation scale and distribution of SMCs risk levels, revealing characteristics of the SMCs risk level distribution. In addition, the number of satellites falling into high risk level at various altitudes and inclinations is presented, which can provide valuable guidance for the targeted deployment of protection strategies for SMCs.