A carregar...
Publicação
A Collaborative Task Scheduling Method for Heterogeneous LEO Satellite Networks Based on Mixed-strategy Game
| Nome: | Descrição: | Tamanho: | Formato: | |
|---|---|---|---|---|
| 767.12 KB | Adobe PDF |
Orientador(es)
Resumo(s)
With the rapid deployment of Low Earth Orbit (LEO) satellite constellations and the increasing demand for satellite-ground collaborative computing, efficient resource management and low-latency processing have become pressing challenges. This paper proposes a novel collaborative planning algorithm, the Evolutionary Reward Function (ERF), which integrates mixed-strategy game theory with Deep Reinforcement Learning (DRL). ERF models the interactive decision-making among heterogeneous satellite nodes through mixed-strategy games and incorporates a dynamic policy adjustment mechanism to enhance task scheduling efficiency, thereby improving mission completion rates in dynamic LEO environments. Extensive simulation results demonstrate that ERF outperforms conventional DRL methods in terms of convergence speed and overall performance.
Descrição
Niu, Q., Teng, H., Jin, S., Wang, Y., Hou, W., & Liu, C. (2026). A Collaborative Task Scheduling Method for Heterogeneous LEO Satellite Networks Based on Mixed-strategy Game. In 2025 International Conference on Satellite Computing, Satellite 2025 Institute of Electrical and Electronics Engineers (IEEE). https://doi.org/10.1109/Satellite67108.2025.11430429
Palavras-chave
Collaborative task scheduling Deep reinforcement learning LEO satellite networks Mixed-strategy game theory Aerospace Engineering Computer Networks and Communications Hardware and Architecture SDG 9 - Industry, Innovation, and Infrastructure
Contexto Educativo
Citação
Editora
Institute of Electrical and Electronics Engineers (IEEE)