A RAG-Enhanced Bi-Level Cognitive Orchestration Framework for LEO Satellite Networks

2026-06-12

AI Summary

This work addresses the challenge of degraded performance in static scheduling for low Earth orbit (LEO) satellite networks, where downlink constraints necessitate on-board computation and induce spatiotemporal resource fragmentation. To overcome this, the study introduces a novel satellite resource orchestration framework that integrates retrieval-augmented generation (RAG) with a dual-layer cognitive architecture. The upper layer dynamically infers multi-objective preference weights by fusing large language models with an expert knowledge base, while the lower layer executes a fidelity-aware genetic algorithm guided by these weights to jointly optimize routing and task offloading. Evaluated on a high-fidelity Walker-Delta constellation simulator, the proposed approach reduces packet loss by 30.7%, improves energy efficiency by 30%, decreases end-to-end latency by 8.5%, and demonstrates robustness under cascading node failures.

Abstract

The rapid growth of remote sensing data in Low Earth Orbit (LEO) satellite networks is increasingly constrained by limited downlink capacity to terrestrial networks. Satellite edge computing alleviates this pressure by enabling in-orbit data processing. However, it introduces a new challenge of spatio-temporal resource fragmentation. Variations in onboard computing capability, constrained energy availability, and intermittent inter-satellite and satellite-ground connectivity lead to highly dynamic and uneven resource distribution, which degrades the performance of conventional static routing and scheduling approaches. To address this, we propose a Retrieval-Augmented Generation (RAG)-enhanced bi-level cognitive orchestration framework for knowledge-guided, multi-objective scheduling. The proposed framework explicitly decouples network control across two different operational scales: at the strategic upper level, a Large Language Model (LLM) leverages an offline-distilled Expert Knowledge Base (EKB) to dynamically infer preference weights based on a compact abstract-state descriptor of real-time network conditions. At the lower execution level, a fidelity-aware genetic scheduler utilizes these inferred weights to compute physically feasible, collision-free joint routing and task offloading schedules. Extensive evaluations on a high-fidelity Walker-Delta network testbed under mixed-criticality workloads demonstrate that the proposed framework effectively consolidates fragmented resources, achieving a 30.7% reduction in packet loss, a 30% improvement in energy efficiency over the most competitive learning-based baseline, and an 8.5% decrease in end-to-end latency, while maintaining robust performance under cascading node-failure scenarios.

Problem

• LEO satellite networks
• resource fragmentation
• edge computing
• dynamic resource distribution
• satellite scheduling

Innovation

• Retrieval-Augmented Generation (RAG)
• Bi-Level Cognitive Orchestration
• LEO Satellite Networks
• Expert Knowledge Base (EKB)
• Fidelity-Aware Genetic Scheduler