A Prediction Model for Channel State Information in Satellite Communication System

As an important complement to sixth-generation (6G) systems, low earth orbit (LEO) satellite will increase system capacity and improve service coverage. A critical question in LEO satellite systems is how to increase spectral efficiency. The existing primary method is the adaptive modulation and coding (AMC) based on accurate channel state information (CSI). However, the long-time delay of LEO satellite will lead to outdated CSI. Existing works usually predicates the future CSI based on time series, which is usually influenced by the constriction on satellite payload, such as nonlinear distortion caused by a high-power amplifier (HPA). In this work, we investigate the effects of predicting CSI using an improved time series prediction model in order to resolve the above problems. The simulation results verify the accuracy of the improved model. Compared to the commonly-used model, the performance of the improved model has a significant increase in spectral efficiency.