Dynamic Multipath Routing Scheme for Delay Differentiated Services in LEO Satellite Networks

Low Earth Orbit (LEO) satellite networks are expected to be seamlessly integrated with terrestrial infrastructures, providing high-quality, globally accessible communication services. Given the diverse service requirements of users, the traffic transmitted through these networks exhibits varying levels of delay tolerance. However, the highly dynamic topology of LEO satellite networks and the limited satellite resources present significant challenges for traditional multipath routing schemes in providing delay differentiated services. To address these issues, this paper proposes a novel Ant Colony Optimization (ACO)-based multipath routing algorithm, which enables each satellite to forward ant packets based on local information to discover paths. This algorithm allows the source satellite to obtain information about multiple disjoint paths that satisfy the delay differentiated service transmission requirements without collecting global information. Furthermore, a dynamic delay differentiated traffic scheduling algorithm is designed to allocate delay-sensitive and delay-tolerant traffic across multiple paths by considering predicted traffic ratios as well as path delay, bandwidth, and energy factors to achieve balanced utilization of network resources. Simulation results show that the proposed routing scheme can provide differentiated services for both types of traffic while reducing both delay and packet loss rate and maintaining balanced utilization of network resources.