Efficient Uplink Transmission in Ultra-Dense LEO Satellite Networks With Multiband Antennas

In this letter, we propose an efficient uplink transmission scheme over ultra-dense low earth orbit (LEO) satellite networks with multiband antennas. A two-timescale transmission resource allocation and frequency band switching optimization framework is established with the objective of maximizing the overall data rate while reducing the band switching cost. The small-timescale transmission resource allocation problem is described as a Markov decision process (MDP) considering continuous LEO satellite movements and Markovian rainfall changes, while the large-timescale satellite band switching is formulated as an integer programming problem. A hierarchical algorithm based on the deep reinforcement learning (DRL) approach and the greedy method is proposed to make efficient decisions. Extensive simulations are conducted to validate the effectiveness and advantages of our proposed scheme.