Joint Beam Direction Control and Radio Resource Allocation in Dynamic Multi-beam LEO Satellite Networks

TL;DR

This paper addresses the challenge of jointly optimizing beam direction and multi-dimensional radio resources in dynamic multi-beam LEO satellite networks to maximize the long-term sum user utility under fairness. It introduces a decomposed framework where beam-direction/time-slot selection and user subchannel assignment are tackled via two-sided matching with externalities, and beam power allocation is solved with successive convex approximation. The key contributions include two swap-based matching algorithms with convergence and stability guarantees, a convexification-based power allocation scheme, and extensive simulations showing up to 2× more served users and up to 68% higher sum user data rate compared to baselines across varied settings. The work demonstrates the practical significance of coordinated beam steering and resource management for improved coverage and efficiency in dynamic LEO constellations.

Abstract

Multi-beam low earth orbit (LEO) satellites are emerging as key components in beyond 5G and 6G to provide global coverage and high data rate. To fully unleash the potential of LEO satellite communication, resource management plays a key role. However, the uneven distribution of users, the coupling of multi-dimensional resources, complex inter-beam interference, and time-varying network topologies all impose significant challenges on effective communication resource management. In this paper, we study the joint optimization of beam direction and the allocation of spectrum, time, and power resource in a dynamic multi-beam LEO satellite network. The objective is to improve long-term user sum data rate while taking user fairness into account. Since the concerned resource management problem is mixed-integer non-convex programming, the problem is decomposed into three subproblems, namely beam direction control and time slot allocation, user subchannel assignment, and beam power allocation. Then, these subproblems are solved iteratively by leveraging matching with externalities and successive convex approximation, and the proposed algorithms are analyzed in terms of stability, convergence, and complexity. Extensive simulations are conducted, and the results demonstrate that our proposal can improve the number of served users by up to two times and the sum user data rate by up to 68%, compared to baseline schemes.

Figures (10)

• Figure 1: An illustration of the studied satellite network scenario.
• Figure 2: Overview of the proposed resource management framework.
• Figure 3: The network topology at the 50-$th$ time slot. The points represent user terminals, the pentagrams represent satellites, and the diamonds represent sub-satellite points.
• Figure 4: The convergence of our proposal with different numbers of beams configured at each satellite ($K=20, K^{thr}=6$).
• Figure 5: Impact of the number of beams configured at each satellite on the network performance ($K=20, K^{thr}=6$). (a): The number of beams vs. sum of $\alpha$-utility; (b) The number of beams vs. sum user data rate; (c) The number of beams vs. the number of served users.

Theorems & Definitions (21)

• Lemma 1
• proof
• Definition 1
• Definition 2
• Proposition 1
• proof
• Definition 3
• Proposition 2
• proof
• Proposition 3