Spatio-temporal correlation-based incomplete time-series traffic prediction for LEO satellite networks

Abstract

Accurate short-term traffic prediction is essential for improving the efficiency of data transmission in low Earth orbit (LEO) satellite networks. However, traffic values may be missing due to collector failures, transmission errors, and memory failures in complex space environments. Incomplete traffic time series prevent the efficient utilization of data, which can significantly reduce the traffic prediction accuracy. To overcome this problem, we propose a novel spatio-temporal correlation-based incomplete time-series traffic prediction (ITP-ST) model, which consists of two phases: reconstituting incomplete time series by missing data imputation and making traffic prediction based on the reconstructed time series. In the first phase, we propose a novel missing data imputation model based on the improved denoising autoencoder (IDAE-MDI). Specifically, we combine DAE with the Gramian angular summation field (GASF) to establish the temporal correlation between different time intervals and extract the structural patterns from the time series. Taking advantage of the unique spatio-temporal correlation of the LEO satellite network traffic, we focus on improving the missing data initialization method for DAE. In the second phase, we propose a traffic prediction model based on a multi-channel attention convolutional neural network (TP-CACNN) by combining the spatio-temporally correlated traffic of the LEO satellite network. Finally, to achieve the ideal structure of these models, we use the multi-verse optimizer (MVO) algorithm to select the optimal combination of model parameters. Experiments show that the ITP-ST model outperforms the baseline models in terms of traffic prediction accuracy at different data missing rates, which demonstrates the effectiveness of our proposed model.

1. Introduction

Due to the high mobility of the low Earth orbit (LEO) satellites, the coverage area of each satellite varies rapidly. Factors such as population size and economic conditions in different coverage areas lead to significant differences in service traffic accessing to the satellites, and dramatic changes in the traffic load can easily lead to network congestion (Jiang et al., 2023). Therefore, it becomes necessary to introduce traffic prediction to sense future traffic changes. However, due to collector failures, transmission errors, and memory failures in the harsh space environment, traffic values may usually be lost at unexpected moments. Such incomplete traffic time series undoubtedly create difficulties in data analysis, which poses a significant challenge for traffic prediction in the LEO satellite networks (Liu W et al., 2021).

In recent decades, efforts have been devoted to solving the traffic prediction problem with incomplete time series. There are two main approaches to address the problem: the two-step approach and the synchronous approach (Wang A et al., 2023). In the former, the missing values in the incomplete time series are first filled in using a missing data imputation method. Then, predictions are made based on the reconstructed complete time series. In the latter, missing data imputation and traffic prediction are synchronized.

Missing data imputation methods are the focus of research in the two-step approach, which generally include mathematical statistics and learning-based methods (Baggag et al., 2021). Commonly used statistical methods include mean interpolation (Mukhopadhyay and Mukherjee, 2020), spline interpolation (You et al., 2020), and the nearest neighbor interpolation (Li et al., 2020). Although these algorithms are simple and easy to implement, these methods are based on simplified smoothness and linearity assumptions, which cannot fully capture the nonlinear and uncertain variations in satellite network traffic and are adapted only to some simple linear time series.