Multi-Path Transmission of Real-Time Remote Sensing Data via Heterogeneous LEO Inter-Satellite-Links

Real-time transmission of remote sensing data is critical for many applications. The rapid enhancement in sensing capabilities has posed challenges in sending large volumes of collected information to users on the ground in a timely manner. Deployed with inter-satellite links (ISLs), low-earth-orbit mega-constellations can facilitate the relay of remote sensing data to ground stations. However, due to scarce spectrum resources, increasing data throughput remains a challenge. Towards this, we explore multi-path transmission of real-time remote sensing data via heterogeneous ISLs, with the aim of maximizing data transmission throughput. We formulate the problem of multi-path transmission via heterogeneous ISLs as a linear program-ming problem, and find that its complexity is exacerbated when dealing with large-scale satellite networks. Therefore, we present an alternative approach using graph theory, transforming the problem into a classical maximum flow problem. This approach ensures optimal resolution of the original problem within polynomial time. Simulations conducted on the Gaofen and Starlink constellation demonstrate the performance benefits of our approach regarding both time complexity and throughput.