Green Traffic Engineering for Satellite Networks Using Segment Routing Flexible Algorithm

2025-09-02

AI Summary

Energy-efficient and reliable routing in large-scale low Earth orbit (LEO) satellite networks remains challenging due to dynamic topologies, stringent latency constraints, and limited onboard power resources. Method: This paper proposes a SRv6-based segment routing traffic engineering framework that jointly optimizes energy efficiency, packet delivery ratio (PDR), and end-to-end latency. It introduces a lightweight SDN control architecture supporting three logically isolated network slices—green (energy-aware), reliable (high-PDR), and real-time (low-latency)—and integrates SRv6’s flexible programmability with mixed-integer linear programming (MILP) optimization and centralized intelligent scheduling, the first such integration for LEO constellations. Contribution/Results: Evaluated on a Telesat Lightspeed constellation simulation, the framework reduces average CPU utilization by 73%, achieves ≥91% PDR under bursty traffic, and cuts end-to-end latency for emergency flows (e.g., Ottawa–Vancouver) by 18 ms, significantly enhancing resource utilization and QoS assurance.

Technology Category

• Planning, Routing, and Scheduling: Optimization of Spatio-temporal Systems
• Search and Optimization: Learning to Search
• Constraint Satisfaction and Optimization: Satisfiability

Application Category

• Systems and Infrastructure for Web, Mobile and WoT: Energy management for devices in mobile Web and WoT environments
• Search and Retrieval-Augmented AI: Efficiency and scalability of Web search engines
• Security and Privacy: Large-scale security measurements

Abstract

Large-scale low-Earth-orbit (LEO) constellations demand routing that simultaneously minimizes energy, guarantees delivery under congestion, and meets latency requirements for time-critical flows. We present a segment routing over IPv6 (SRv6) flexible algorithm (Flex-Algo) framework that consists of three logical slices: an energy-efficient slice (Algo 130), a high-reliability slice (Algo 129), and a latency-sensitive slice (Algo 128). The framework provides a unified mixed-integer linear program (MILP) that combines satellite CPU power, packet delivery rate (PDR), and end-to-end latency into a single objective, allowing a lightweight software-defined network (SDN) controller to steer traffic from the source node. Emulation of Telesat's Lightspeed constellation shows that, compared with different routing schemes, the proposed design reduces the average CPU usage by 73%, maintains a PDR above 91% during traffic bursts, and decreases urgent flow delay by 18 ms between Ottawa and Vancouver. The results confirm Flex-Algo's value as a slice-based traffic engineering (TE) tool for resource-constrained satellite networks.

Problem

• Minimizing energy consumption in satellite network routing
• Ensuring reliable packet delivery under network congestion
• Meeting latency requirements for time-critical data flows

Innovation

• SRv6 Flex-Algo framework with three logical slices
• Unified MILP combining power, delivery rate, latency
• Lightweight SDN controller for source traffic steering