Author(s): E Venkatesan and V Thangavel

Abstract: Satellite communication systems play a pivotal role in global connectivity, yet they remain vulnerable to failures caused by hardware degradation, environmental disturbances, and ground-segment errors. This study proposes a structured methodology to identify, isolate, and recover from such failures using intelligent algorithmic approaches. The methodology begins with system modeling, simulating satellite subsystems and environmental factors under both nominal and fault conditions. Fault Detection and Isolation (FDI) techniques are applied to telemetry data to detect anomalies and locate the source of faults. To enhance recovery, four approaches are evaluated: decision-tree models for structured fault classification, Bayesian inference for probabilistic reasoning under uncertainty, self-healing protocols for autonomous subsystem reconfiguration, and a hybrid framework that integrates multiple methods. Performance is assessed using metrics such as detection accuracy, false alarm rate, recovery time, and link availability. Results demonstrate that individual algorithms provide specific advantages, whereas hybrid models offer superior resilience by combining deterministic, probabilistic, and self-healing strategies. The proposed methodology highlights the potential of intelligent algorithmic solutions to improve the autonomy, reliability, and robustness of satellite communication networks, providing a foundation for more resilient and self-sustaining space systems.

DOI: 10.33545/27075923.2025.v6.i2b.109

Pages: 96-101 | Views: 204 | Downloads: 138

Download Full Article: Click Here