Constellation Topology Design for Maximum Capacity of LEO Satellite Networks

The low-earth-orbit (LEO) satellite constellation networks play a vital role due to their potential to provide high throughput in response to the escalating demands of future communication networks. However, inadequate matching between the constellation topology and traffic distribution would result in network congestion, which degrades the throughput of the LEO network. In this paper, we aim to enhance the throughput of LEO networks through constellation design. Especially, to provide a theoretical guideline for topology design, we prove that the achievable throughput capacity upper bound equals <inline-formula xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink"> <tex-math notation="LaTeX">$3\sqrt {2N}\left ({{W_{lx}+W_{ly}}}\right)$ </tex-math></inline-formula>, where N represents the constellation size, and <inline-formula xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink"> <tex-math notation="LaTeX">$W_{lx}$ </tex-math></inline-formula> and <inline-formula xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink"> <tex-math notation="LaTeX">$W_{ly}$ </tex-math></inline-formula> represent the inter- and intra-plane data rates of the inter-satellite link (ISL), respectively. Aided by this, a throughput capacity maximum topology design (TCMTD) algorithm is proposed to determine the constellation parameters and connection relationships. Consequently, the average path length of the data packet transmission can be minimized and the utilization rate of each ISL can be maximized, thereby achieving the throughput capacity upper bound. Furthermore, a throughput capacity enhanced topology design (TCETD) algorithm is proposed to achieve a near-optimal throughput when some ISLs cannot be constructed due to LoS constraints. Both algorithms take into account the constraints including phase factors and LoS constraints. Simulation results conducted using OPNET demonstrate the effectiveness of the proposed algorithms.