Satellite communication systems are vulnerable to signal degradation caused by atmospheric conditions such as rain, snow, and fog. In particular, rain fade is a major challenge that is responsible for significant signal attenuation and degradation. Site diversity techniques have been used to mitigate the effects of rain fade by transmitting the same data through different locations and combining the received signals to reduce errors. However, the effectiveness of site diversity depends on having accurate information about the instantaneous values and evolution of rain attenuation at each of the cooperating diversity sites. We propose using deep learning techniques to predict rain fade when signal reception drops to a pre-set threshold in diversity. We describe the use of deep learning models such as the Gated Recurrent Unit (GRU), a recurrent neural network (RNN) to predict rain fade based on immediate recent past fade data captured at the Chilbolton Observatory and Chilton, both in southern England. To demonstrate the effectiveness of our proposed model, we perform experiments on a real-world dataset of satellite communication beacon signals and compare the results with those obtained without using any prediction method. Our results show that the proposed deep learning models outperform the traditional models in terms of reliability, accuracy and precision. Furthermore, we looked at failure points, where none of the sites is capable of delivering a steady link even with the use of this AI technique. In conclusion, our proposed deep learning approach offers a promising solution for predicting satellite signal fade caused by rain, which could greatly improve the reliability and efficiency of satellite communication systems.
AI-optimised site diversity switching for geostationary satellite links
2023-01-01
291181 byte
Conference paper
Electronic Resource
English
Elliptic Satellite Replacements for Geostationary Satellite Systems with 80/140 GHz Links
British Library Conference Proceedings | 2009
|Channel Adaptive Satellite Diversity for Non-Geostationary Mobile Satellite Systems
British Library Conference Proceedings | 1997
|