Resource allocation strategy is drawing much more attention in satellite networks due to limited resources, e.g., limited observation resources and transmission resources. For earth observation tasks, not all potential observation tasks are being observed, and not all observation data can be transmitted to the ground station. It is especially crucial to improve the efficiency of resource utilization and satisfy the Quality of Service (QoS) requirements of users. In this paper, we try to implement matching theory to solve the resource allocation problem. We propose a novel one-to-one matching model under bilateral preferences by building the preference functions for tasks and resources. Then, we introduce the Task-Oriented Gale-Shapley (T-O GS) algorithm and Adjacent Time Slot Matching (ATSM) algorithm to dynamically achieve stable matching. Furthermore, we analyze the stability and uniqueness of the proposed algorithm. The simulation results demonstrate that the proposed algorithm can significantly improve the availability and the efficiency of resource allocation with low computational complexity.

由于资源有限，例如有限的观测资源和传输资源，资源分配策略在卫星网络中越来越受到关注。对于地球观测任务，并不是所有潜在的观测任务都被观测到，也不是所有的观测数据都可以传输到地面站。提高资源利用效率，满足用户的服务质量(QoS)要求尤为关键。在本文中，我们尝试实施匹配理论来解决资源分配问题。我们通过构建任务和资源的偏好函数，在双边偏好下提出了一种新颖的一对一匹配模型。然后，我们引入了面向任务的 Gale-Shapley (TO GS) 算法和相邻时隙匹配 (ATSM) 算法来动态实现稳定匹配。此外，我们分析了所提出算法的稳定性和唯一性。仿真结果表明，该算法能够显着提高资源分配的可用性和效率，且计算复杂度较低。