Jointly Optimizing Satellite Handover and Power Allocation in LEO Satellite Network: A Dual-Agent Framework
IEEE Transactions on Vehicular Technology · 2026-08 · 75 (8) · 18614-18619
DOI: 10.1109/tvt.2026.3674597 ↗Kurz gefasst
Fachartikel in der Zeitschrift IEEE Transactions on Vehicular Technology, 2026-08. Verfasst von Qingyuan Zhang und 2 weiteren Autor:innen. Die Arbeit wurde bisher 4-mal zitiert (Crossref) und nennt 16 Literaturangaben. Gefördert von: National Key Laboratory of Advanced Communication Networks Foundation Project, National Natural Science Foundation of China.
Themen: Übergabe
Abstract (Originalsprache)
Low-earth orbit (LEO) satellites are crucial for next-generation wireless networks, providing low-latency communication and wide coverage. However, high-speed mobility and energy limits cause frequent handovers and power allocation issues. Existing works often rely on single-agent models, which struggles to jointly optimize handover decisions with discrete variable and power allocation with continuous variable. To address this issue, we propose a dynamic dual-agent optimization framework (D2AOF) to optimize both handovers and power allocation in LEO satellite networks. Specifically, D2AOF combines deep Q-network (DQN) for optimizing handover decisions and proximal policy optimization (PPO) for refining power allocation, as an integrated iterative intelligent framework. The DQN agent dynamically optimizes handover decisions, ensuring the users are connected to the most suitable satellites. The PPO agent refines power allocation to improve the resource utility. Our optimization goal is to maximize the number of users completing data transmission within the specified time limit, while ensuring optimal transmission rates for connected users in each time slot. Therefore, we introduce a multi-dimensional reward function, incorporating user throughput and the number of timed-out users to balance resource allocation. Simulation results show that D2AOF achieves a 16% higher reward than an alternative dual-agent model and a 40% improvement over single-agent models, while boosting average throughput by 13% and reducing timed-out users by 30%.
Förderung
- National Key Laboratory of Advanced Communication Networks Foundation Project
- National Natural Science Foundation of China
Zitieren
APA
Zhang, Q., Fu, S., & Yang, Z. (2026). Jointly Optimizing Satellite Handover and Power Allocation in LEO Satellite Network: A Dual-Agent Framework. IEEE Transactions on Vehicular Technology, 75(8), 18614-18619. https://doi.org/10.1109/tvt.2026.3674597
Vancouver
Zhang Q, Fu S, Yang Z. Jointly Optimizing Satellite Handover and Power Allocation in LEO Satellite Network: A Dual-Agent Framework. IEEE Transactions on Vehicular Technology. 2026;75(8):18614-18619. doi:10.1109/tvt.2026.3674597
BibTeX
@article{Zhang2026,
author = {Zhang, Qingyuan and Fu, Shu and Yang, Zhaohui},
title = {Jointly Optimizing Satellite Handover and Power Allocation in LEO Satellite Network: A Dual-Agent Framework},
journal = {IEEE Transactions on Vehicular Technology},
year = {2026},
volume = {75},
number = {8},
pages = {18614--18619},
doi = {10.1109/tvt.2026.3674597}
}
EUVIMED