Adaptive Timescale Load Balancing Routing Algorithm for LEO Satellite Network

The LEO satellite network demonstrates excellent communication capabilities, including uninterrupted service provision, large bandwidth, low delay, and wide geographic coverage, highlighting its important role in global connectivity. Due to the uneven distribution of global traffic, the links between developed regions are more prone to congestion, while other links are relatively idle. In order to improve the throughput of the network while ensuring good end-to-end delay performance, an adaptive timescale load balancing (ATLB) routing algorithm based on SDN is proposed in this paper. By modeling the link historical traffic through an approach based on empirical mode decomposition, ATLB classifies congestion as transient and persistent. When a transient congestion occurs, the pre-configured alternative forwarding path will work for local load balancing. And when persistent congestion occurs, ATLB will perform global load balancing on the appropriate timescale determined by the control plane adaptively. The simulation results show that ATLB has better comprehensive performance in packet drop rate, end-to-end delay and throughput compared with other algorithms.