AI-Augmented SysML Framework for Early Failure Mitigation in Small Satellite Communications: Link-Budget Verification and Trade Study Integration

Abstract

Multi-billion-dollar losses in small satellite programs have been linked to early communication failures and unverified link budgets. Ensuring reliable communication from the start has become a central difficulty for aerospace projects. When faults occur, missions are delayed, confidence in the program is damaged, and weaknesses in conventional approaches are revealed. In many cases, RF margin problems are discovered late in the lifecycle, when redesigns are costly. The challenge is intensified by short schedules, disconnected engineering tools, and limited resources for thorough verification in the early design phase. This paper presents a model-based systems engineering approach that incorporates artificial intelligence to address these risks. The framework connects SysML requirement models with executable link-budget analysis and extends this link to multi-criteria trade studies across UHF, S-band, and X-band radio systems. Artificial intelligence techniques allow rapid ranking of subsystem choices using cost, efficiency, and reliability as evaluation factors. SysML provides direct traceability, linking requirements to verification steps and to design decisions. The study shows that integrating link-budget checks and AI-supported trade studies within SysML models detects margin issues early, reduces redesigned cycles, and brings greater clarity to subsystem selection. The approach demonstrates how combining automated decision analysis with model-based traceability lowers communication risks and improves lifecycle performance. Main contributions include a SysML-based link-budget verification method with AI support, an AI-driven trade study process for subsystem evaluation, and a tested approach that strengthens reliability, traceability, and efficiency in small satellite communication design.