Multi-User Computation Offloading in Dynamic Environment for LEO Constellation Networks

• Published in: IEEE transactions on vehicular technology Vol. 73; no. 3; pp. 4453 - 4458
• Main Authors: Gao, Yufang, Zhao, Kanglian, de Cola, Tomaso, Li, Wenfeng, Cui, Yan, Hu, Peicong
• Format: Journal Article
• Language: English
• Published: New York IEEE 01.03.2024; The Institute of Electrical and Electronics Engineers, Inc. (IEEE)
• Subjects: Algorithms; Cochannel interference; Computation offloading; Computational modeling; dynamic environment; Game theory; Games; Interference; Low Earth Orbit (LEO) constellation network; Low earth orbit satellites; Low earth orbits; Machine learning; Performance degradation; potential game; Satellites; Task analysis; Vehicle dynamics
• ISSN: 0018-9545; 1939-9359
• DOI: 10.1109/TVT.2023.3327308

In this paper, we examine the problem of multi-user computation offloading for Low Earth Orbit (LEO) constellation networks under dynamic environment. To solve the puzzle of in-orbit computing performance degradation caused by multi-user access interference, we focus on minimizing channel interference while considering the stochastic characteristic of satellite-ground channels and the dynamic nature of ground user access. From a game-theoretic perspective, we model the interference mitigation problem as a potential game, and demonstrate that the Nash Equilibrium (NE) minimizes the channel interference. Meanwhile, we design an intelligent stochastic learning algorithm (ISLA) based on the stochastic learning method. Finally, simulation results show that the proposed algorithm can significantly reduce co-channel interference and system overhead compared with the benchmark algorithms, and achieve near-optimal performance.