Multi-beam satellite communication systems have been received widespread attention due to their high throughput and efficient resource utilization. In this paper, we investigate the beam illumination and resource allocation problem in multi-beam satellite communication systems. By jointly considering user position and service characteristics, an optics-based initial user grouping algorithm is proposed. To enhance beam coverage performance, a minimum circle algorithm is proposed to optimally design satellite beam positions and coverage radius. Given the obtained user grouping strategy, we address the difference between random user service demands and service provisioning capability of the system, and define system cost function. The joint beam illumination, sub-channel and power allocation problem is formulated as a system cost function minimization problem. To solve the formulated optimization problem, we introduce aggregate nodes to describe the characteristics of user groups, and address the beam illumination and power allocation problem of user groups. The problem is modeled as a mixed-space Markov decision process (MDP), and a parameterized deep Q-network-based joint beam illumination and power allocation algorithm is proposed. Based on the obtained resource allocation strategy for user groups, we then design user-oriented sub-channel and power allocation strategy. To this end, we model the optimization problem as an MDP and propose a double deep Q-network (DDQN) algorithm-based algorithm. To address the concern that the DDQN algorithm may reach a local optimum, proximal policy optimization algorithms with discrete action space and continuous action space are proposed. Simulation results validate the effectiveness of the proposed algorithms.

中文翻译：

---

多波束卫星通信系统的动态资源分配


多波束卫星通信系统因其高吞吐量和高效的资源利用而受到广泛关注。在本文中，我们研究了多波束卫星通信系统中的波束照明和资源分配问题。综合考虑用户位置和服务特征，提出一种基于光学的初始用户分组算法。为了增强波束覆盖性能，提出了最小圆算法来优化设计卫星波束位置和覆盖半径。给定获得的用户分组策略，我们解决随机用户服务需求与系统的服务提供能力之间的差异，并定义系统成本函数。联合光束照明、子信道和功率分配问题被表述为系统成本函数最小化问题。为了解决公式化的优化问题，我们引入聚合节点来描述用户组的特征，并解决用户组的光束照明和功率分配问题。该问题被建模为混合空间马尔可夫决策过程（MDP），并提出了一种基于参数化深度Q网络的联合光束照明和功率分配算法。根据获得的用户组资源分配策略，设计面向用户的子信道和功率分配策略。为此，我们将优化问题建模为 MDP，并提出一种基于双深度 Q 网络（DDQN）算法的算法。为了解决DDQN算法可能达到局部最优的问题，提出了离散动作空间和连续动作空间的近端策略优化算法。仿真结果验证了所提算法的有效性。