Dynamic collaborative data download in heterogeneous satellite networks

Low-earth-orbit (LEO) satellite network has become a critical component of the satellite-terrestrial integrated network (STIN) due to its superior signal quality and minimal communication latency. However, the highly dynamic nature of LEO satellites leads to limited and rapidly varying contact time between them and Earth stations (ESs), making it difficult to timely download massive communication and remote sensing data within the limited time window. To address this challenge in heterogeneous satellite networks with coexisting geostationary-earth-orbit (GEO) and LEO satellites, this paper proposes a dynamic collaborative inter-satellite data download strategy to optimize the long-term weighted energy consumption and data downloads within the constraints of on-board power, backlog stability and time-varying contact. Specifically, the Lyapunov optimization theory is applied to transform the long-term stochastic optimization problem, subject to time-varying contact time and on-board power constraints, into multiple deterministic single time slot problems, based on which online distributed algorithms are developed to enable each satellite to independently obtain the transmit power allocation and data processing decisions in closed-form. Finally, the simulation results demonstrate the superiority of the proposed scheme over benchmarks, e.g., achieving asymptotic optimality of the weighted energy consumption and data downloads, while maintaining stability of the on-board backlog.