A Prediction-Based Resource Matching Scheme for Rentable LEO Satellite Communication Network

The resource allocation strategy is a critical problem in the low earth orbit (LEO) satellite communication network (SCN) due to the limited resources and high movement. In this letter, we first introduce the rental service into LEO-SCN to improve the resource utilization by renting out unused resources. Secondly, we investigate a prediction-based resource matching scheme for rentable LEO-SCN. Specifically, a long-short-term memory (LSTM) based prediction framework is developed to predict the traffic value of local tasks, which is affected by the periodic motion and the inherent traffic law. Then, based on the predicted traffic, a task-driven joint resource allocation method is proposed to efficiently match power and spectrum resources. Finally, the simulation results verify the validity of the proposed approach compared with the pre-allocation strategy and the adaptive allocation strategy. Moreover, we analyze the key factors that affect the achievable performance of the proposed prediction-based resource matching scheme.