Deep Learning Based Cell Reselection Trigger Optimization in Satellite Communication Networks

Due to the characteristics of low Earth orbit (LEO) satellite networks, such as high satellite mobility, large network scale, and uneven user distribution, efficient mobility management is crucial to ensure the continuity and stability of user services. Traditional mobility management schemes are specifically designed for terrestrial cellular networks, making it difficult for them to adapt to the large scale and highly dynamic characteristics of LEO satellite networks. This paper focuses on proposing a efficient mobility management for users in non-connected state in LEO satellite networks. It leverages machine learning algorithms to develop a cell reselection evaluation trigger mechanism aimed at reducing reselection frequency and minimizing unnecessary evaluation triggers, thereby ensuring service quality for users while improving system efficiency.