Article
Engineering, Electrical & Electronic
Zhiyuan Lin, Zuyao Ni, Linling Kuang, Chunxiao Jiang, Zhen Huang
Summary: This paper proposes a dynamic beam pattern and bandwidth allocation scheme based on deep reinforcement learning (DRL) and introduces a cooperative multi-agents deep reinforcement learning (MADRL) framework to solve the joint allocation problem of bandwidth and beam pattern. Simulation results demonstrate that the proposed method can adapt to non-uniform and time-varying traffic demands and has good generalization ability.
IEEE TRANSACTIONS ON VEHICULAR TECHNOLOGY
(2022)
Article
Engineering, Civil
Ying He, Yuhang Wang, F. Richard Yu, Qiuzhen Lin, Jianqiang Li, Victor C. M. Leung
Summary: This paper explores the use of low orbit multi-beam satellite-terrestrial networks to serve vehicles. A multi-agent actor-critic method with attention mechanism is proposed for resource allocation to vehicles with strict delay requirements and minimum bandwidth consumption. Extensive simulation results verify the effectiveness of the proposed method in achieving efficient resource allocation on-demand for vehicles under strictly limited bandwidth resources.
IEEE TRANSACTIONS ON INTELLIGENT TRANSPORTATION SYSTEMS
(2022)
Article
Telecommunications
Jili Wang, Bangning Zhang, Bing Zhao, Guoru Ding, Daoxing Guo
Summary: This study investigates the bandwidth allocation problem for discontinuous frequency bands in cognitive satellite-terrestrial networks (CSTNs) and proposes a solution based on a bandwidth auction game, along with an improved learning algorithm to achieve satisfactory performance.
IEEE COMMUNICATIONS LETTERS
(2021)
Article
Computer Science, Artificial Intelligence
Luona Wei, Yuning Chen, Ming Chen, Yingwu Chen
Summary: The paper proposes a deep reinforcement learning and parameter transfer based approach (RLPT) for tackling the multi-objective version of the agile earth observation satellite scheduling problem (MO-AEOSSP). RLPT decomposes the problem into scalarized sub-problems and applies neural networks and reinforcement learning to generate high-quality schedules. Experimental results show that RLPT outperforms traditional algorithms in terms of solution quality, distribution, and computational efficiency.
APPLIED SOFT COMPUTING
(2021)
Article
Engineering, Aerospace
Changhao Sun, Xiaochu Wang, Huaxin Qiu, Qingrui Zhou
Summary: This paper introduces a distributed task allocation algorithm via learning in games for multi-satellite systems, which converges to near optimal assignments with probability 1 in finite time and demonstrates robustness against incidental disturbances. Numerical experiments show the positive effect of memory length on solution efficiency refinement, while comparison experiments highlight the superiority of the presented methodology over existing methods.
AEROSPACE SCIENCE AND TECHNOLOGY
(2021)
Article
Telecommunications
Flor G. Ortiz-Gomez, Daniele Tarchi, Ramon Martinez, Alessandro Vanelli-Coralli, Miguel A. Salas-Natera, Salvador Landeros-Ayala
Summary: This paper discusses the use of reinforcement learning algorithms and deep reinforcement learning algorithms to manage resources in flexible payload architectures in very high throughput satellite systems. The performance and complexity of these algorithms are compared, and the superiority of cooperative multiagent distribution is demonstrated.
IEEE TRANSACTIONS ON COGNITIVE COMMUNICATIONS AND NETWORKING
(2022)
Article
Computer Science, Information Systems
Akito Suzuki, Ryoichi Kawahara, Shigeaki Harada
Summary: This paper proposes a dynamic virtual network allocation method based on cooperative multi-agent deep reinforcement learning, which can quickly optimize network resources even when network demands change drastically. The results show that this method can calculate effective allocation within 1 second, reducing server and link utilization and constraint violations significantly.
IEEE TRANSACTIONS ON NETWORK AND SERVICE MANAGEMENT
(2022)
Article
Computer Science, Artificial Intelligence
Amirhossein Asgharnia, Howard Schwartz, Mohamed Atia
Summary: This paper proposes a fuzzy multi-objective reinforcement learning algorithm that can solve multi-objective problems and handle continuous state-action domains.
Article
Engineering, Electrical & Electronic
Tedros Salih Abdu, Steven Kisseleff, Eva Lagunas, Symeon Chatzinotas
Summary: Conventional geostationary satellite communication systems with uniform resource allocation are inefficient in the presence of non-uniform demand distribution. The next generation of broadband satellite systems will enable flexibility in resource allocation. A novel satellite resource assignment design aims to maximize satellite spectrum utilization, and its efficiency is validated through an effective solution.
IEEE TRANSACTIONS ON WIRELESS COMMUNICATIONS
(2021)
Article
Computer Science, Artificial Intelligence
Junwei Ou, Lining Xing, Feng Yao, Mengjun Li, Jimin Lv, Yongming He, Yanjie Song, Jian Wu, Guoting Zhang
Summary: This article proposes a solution to the satellite range scheduling problem by combining deep reinforcement learning with a heuristic scheduling method. The algorithm decomposes the problem into two subproblems: task assignment and single antenna scheduling. The DRL is used to determine the task assignment process, and the heuristic scheduling method is utilized to solve the single antenna scheduling problem quickly. Experimental results demonstrate that this method effectively deals with the satellite range scheduling problem.
SWARM AND EVOLUTIONARY COMPUTATION
(2023)
Article
Computer Science, Information Systems
Wenyu Sun, Weijia Zhang, Ning Ma, Min Jia
Summary: In this paper, a reinforcement learning-based Multi-Branch Deep Q-Network (MBDQN) is proposed to improve the resource occupancy of satellite communication systems. MBDQN extracts features of the satellite resource pool state and task state simultaneously and calculates the action-value function. Experimental results show that our proposed method achieves superior performance compared to greedy or heuristic methods on the generated task datasets.
Article
Telecommunications
Ailing Xiao, Zhenming Chen, Sheng Wu, Shichao Jin, Li Ma
Summary: The proposed strategy for beam-hopping low-earth-orbit satellites utilizes long-term and short-term bandwidth allocation methods, effectively improving communication efficiency and reducing blocking ratio.
IEEE COMMUNICATIONS LETTERS
(2022)
Article
Computer Science, Artificial Intelligence
Xianglong Li, Yuan Li, Jieyuan Zhang, Xinhai Xu, Donghong Liu
Summary: This paper proposes a hierarchical framework for learning the dynamic allocation and cooperative behaviors of agents, achieving better performance than other methods in challenging environments.
COMPLEX & INTELLIGENT SYSTEMS
(2023)
Article
Engineering, Aerospace
Xiaokai Zhang, Bangning Zhang, Daoxing Guo, Kang An, Shuai Qi, Gang Wu
Summary: This article introduces a potential game-based approach for collaborative user scheduling and power allocation in uplink multibeam satellite-based Internet of things networks. The framework, optimization problem, game-theoretic model, and iterative algorithm are proposed and implemented to achieve the goal effectively. Simulation results demonstrate the convergence and effectiveness of the proposed approach.
IEEE TRANSACTIONS ON AEROSPACE AND ELECTRONIC SYSTEMS
(2021)
Article
Computer Science, Information Systems
Sriharsha Chigullapally, C. Siva Ram Murthy
Summary: In this paper, we propose a UAV-enabled MEC service for connecting IoRT devices to the internet. The data collected by UAVs is transmitted to ground MEC devices, which then send back results to the UAVs and relay them to the IoRT devices. The problem is formulated as a complex non-convex optimization problem and solved using an iterative algorithm. Numerical results show the superiority of our proposed model in terms of energy consumption and throughput.
COMPUTER COMMUNICATIONS
(2023)