International Journal of Scientific Research and Engineering Development

Satellite communication systems operating in Ku-, Ka-, Q-, and V-bands are highly vulnerable to rain attenuation, which degrades signal quality, increases BER, reduces SNR, and lowers link availability. Conventional propagation models such as ITU-R and Crane provide long-term attenuation estimates but fail to accurately predict short-term atmospheric variations. Recent advances in artificial intelligence have enabled the development of machine learning and deep learning models for real-time rain attenuation forecasting and adaptive fade mitigation. This survey reviews attenuation mechanisms, traditional prediction methods, and advanced deep learning architectures including LSTM, CNN, GRU, BiLSTM, hybrid CNNLSTM, and transformer-based models. It also examines satellite link reliability prediction using RSSI and Doppler parameters, intelligent NTN communication systems, MATLAB-based implementation approaches, performance evaluation metrics, current research challenges, and future directions. The study demonstrates that deep learning significantly enhances attenuation prediction accuracy, communication reliability, and adaptive satellite network performance under severe weather conditions.

Ameen Ali, Dr Prabhat Sharma,"Deep Learning-Based Rain Attenuation Prediction and Satellite Link Reliability Enhancement in High-Frequency Satellite Communication Systems: A Comprehensive Survey" International Journal of Scientific Research and Engineering Development, V9(3): Page(2190-2198) May-June 2026. ISSN: 2581-7175. www.ijsred.com. Published by Scientific and Academic Research Publishing.