AI-driven Satellite Network Routing and Planning

Efficient network planning and routing are critical for the optimal performance of satellite communication systems. Traditional satellite communication does not consider atmospheric and path loss in routing decisions due to atmospheric uncertainties of Earth to satellite and satellite to Earth and dynamic changes in satellite network topology. This work proposed a novel method to predict atmospheric and path loss in satellite networks by integrating Artificial Intelligence (AI) techniques. Experimental results demonstrate that the proposed AI-DVTR algorithm achieves higher accuracy for predicting atmospheric losses than the traditional DV-DVTR algorithm. Specifically, AI-DVTR enhances path loss efficiency by 80% compared to the DV-DVTR algorithm.