Completion Time Optimization with Coupled Uplink-Downlink Resource Allocation for Satellite Systems

Satellites with regenerative payload have become an important way to decrease the end-to-end data transmission latency and improve the resource utilization efficiency in satellite communication systems. However, most of existing works focus on individual uplink or downlink resource allocation without considering the coupling effects of uplink and downlink resource management. In this paper, the coupled uplink-downlink resource allocation with regenerative payloads is analyzed and formulated as a mixed-integer nonlinear programming problem to minimize the completion time of data transmission. To solve the complex problem, a two-stage algorithm is proposed. In the first stage, deep reinforcement learning is adopted to make decisions on the paired user scheduling and channel allocation for uplink and downlink transmission. With the fixed channel allocation decisions, the optimal downlink power allocation can be obtained via successive convex approximation at the second stage. Moreover, the proposed two-stage algorithm can be implemented with low computation complexity. Simulation results show that the proposed algorithm can effectively reduce the completion time of end-to-end data transmission. IEEE