ISBN: 978-981-18-7950-0 DOI: 10.18178/wcse.2023.06.001
A Hidden Markov Model Based Dynamic Power and Frequency Joint Allocation Scheme for GEO Satellite System
Abstract—This paper studies the power and frequency allocation method for downlink data transmission of GEO satellites with multi-beam. An HMM-DPFJA scheme is proposed, which introduces an HMM based prediction network to forecast the traffic requirement, and adopting a deep reinforce learning based scheme to pre-allocate power and frequency resources. The simulation results shows that HMM-DPFJA scheme can successfully pre-allocate power and frequency resources before traffic requirement variation happens, and is greatly superior to uniform allocation scheme in the region that traffic requirement is high.
Index Terms—satellite system, resource allocation, reinforce learning, hidden Markov model
Ben Wang
Nanjing University of Aeronautics and Astronautics, CHINA
Nanjing Marine Radar Institute, China State Shipbuilding Corporation, CHINA
Yuanyuan Xu
Beijing University of Posts and Telecommunications, CHINA
Yue Yuan, Lin Fu
Nanjing Marine Radar Institute, China State Shipbuilding Corporation, CHINA
Lin Xu
Nanjing University of Aeronautics and Astronautics, CHINA
Cite: Ben Wang, Yuanyuan Xu, Yue Yuan, Lin Fu, Lin Xu, "A Hidden Markov Model Based Dynamic Power and Frequency Joint Allocation Scheme for GEO Satellite System " Proceedings of 2023 the 13th International Workshop on Computer Science and Engineering (WCSE 2023), pp. 1-6, June 16-18, 2023.