Carbon-Aware Orchestration of Integrated Satellite Aerial Terrestrial Networks via Digital Twin

TL;DR

This work addresses the carbon footprint of integrated satellite-aerial-terrestrial networks (ISATNs) by introducing a Digital Twin–driven, carbon-aware orchestration framework that uses $gCO_2e/bit$ as a primary sustainability metric. It combines a day-ahead MPC planner with online RL for real-time control within a PDCA loop, enabling proactive forecasting, scenario evaluation, and rapid corrective actions. The framework leverages ISATN-specific controls such as carbon-aware handovers, UAV duty-cycling, and renewable-aware edge placement, achieving up to 29% reductions in $gCO_2e/bit$ versus QoS-only orchestration and improving renewable utilization to 58% with enhanced resilience to rain events and traffic surges. These results demonstrate that sustainability objectives can be met without compromising QoS, highlighting the DT-enabled approach as a practical pathway toward greener 6G ISATNs and informing future multi-domain interoperability and lifecycle emissions research.

Abstract

Integrated Satellite Aerial Terrestrial Networks (ISATNs) are envisioned as key enablers of 6G, providing global connectivity for applications such as autonomous transportation, Industrial IoT, and disaster response. Their large-scale deployment, however, risks unsustainable energy use and carbon emissions. This work advances prior energy-aware studies by proposing a carbon-aware orchestration framework for ISATNs that leverages Digital Twin (DT) technology. The framework adopts grams of CO$_2$-equivalent per bit (gCO$_2$/bit) as a primary sustainability metric and implements a multi timescale Plan Do Check Act (PDCA) loop that combines day-ahead forecasting with real-time adaptive optimization. ISATN-specific control knobs, including carbon-aware handovers, UAV duty cycling, and renewable-aware edge placement, are exploited to reduce emissions. Simulation results with real carbon intensity data show up to 29\% lower gCO$_2$/bit than QoS-only orchestration, while improving renewable utilization and resilience under adverse events.

Figures (5)

• Figure 1: Carbon-aware orchestration framework for ISATNs. Telemetry informs DT planning and adjustment, the orchestrator generates policies, and actuators execute them, aligning with the PDCA cycle (i.e., Plan, Do, Check, and Act).
• Figure 2: Dual-timescale carbon-aware orchestration in ISATNs. Day-ahead MPC generates a baseline plan over hours, while the real-time RL agent performs fast adjustments over seconds. Both are supported by the Digital Twin for forecasting, validation, and telemetry feedback.
• Figure 3: Carbon intensity over 7 days with emissions per GB for different strategies. The proposed MPC+RL orchestration closely aligns traffic with low-carbon periods.
• Figure 4: Weekly energy consumption by network layer. MPC+RL orchestration lowers RAN idle and satellite energy while shifting workloads to renewable-rich nodes.
• Figure 5: p95 latency during the Day 3 rain event. MPC+RL orchestration stabilizes within 60 seconds, outperforming other strategies.