[Paper Review] Satellite edge computing for real-time and very-high resolution Earth observation

Israel Leyva‐Mayorga, Marc Martinez-Gost+5 more|VBN Forskningsportal (Aalborg Universitet)|Dec 25, 2022

Satellite Communication Systems4 citations

TL;DR

This paper proposes a satellite mobile edge computing (SMEC) framework that leverages inter-satellite links (ISLs) to distribute image processing and compression tasks across Low Earth Orbit (LEO) satellites, minimizing energy consumption while maximizing real-time, high-resolution Earth observation. By jointly optimizing data distribution, compression ratios, and CPU frequencies, the framework increases system capacity by 12× over direct downlink and reduces energy use by up to 90% in optimized scenarios.

ABSTRACT

In real-time and high-resolution Earth observation imagery, Low Earth Orbit (LEO) satellites capture images that are subsequently transmitted to ground to create an updated map of an area of interest. Such maps provide valuable information for meteorology or environmental monitoring, but can also be employed in near-real time operation for disaster detection, identification, and management. However, the amount of data generated by these applications can easily exceed the communication capabilities of LEO satellites, leading to congestion and packet dropping. To avoid these problems, the Inter-Satellite Links (ISLs) can be used to distribute the data among the satellites for processing. In this paper, we address an energy minimization problem based on a general satellite mobile edge computing (SMEC) framework for real-time and very-high resolution Earth observation. Our results illustrate that the optimal allocation of data and selection of the compression parameters increase the amount of images that the system can support by a factor of 12 when compared to directly downloading the data. Further, energy savings greater than 11% were observed in a real-life scenario of imaging a volcanic island, while a sensitivity analysis of the image acquisition process demonstrates that potential energy savings can be as high as 92%.

Motivation & Objective

• Address the growing data congestion in high-resolution Earth observation from LEO satellites due to limited downlink capacity.
• Overcome the limitations of direct ground station downlink by enabling distributed processing across a satellite constellation.
• Minimize energy consumption in satellite processing and communication while maintaining real-time performance and high image resolution.
• Optimize the joint allocation of data, compression ratios, and CPU frequencies across satellites to improve system efficiency.
• Enable feasible real-time monitoring of dynamic events such as volcanic eruptions through energy-efficient, distributed processing.

Proposed method

• Formulate a satellite mobile edge computing (SMEC) framework using inter-satellite links (ISLs) to offload processing and reduce downlink load.
• Model the system as a joint optimization problem minimizing energy consumption subject to latency, CPU, and transmission constraints.
• Introduce a time-division-based task scheduling model where each satellite processes frames with optimized compression ratios ρk and CPU frequencies f(n)k.
• Use a Lagrangian dual decomposition method to solve the non-convex optimization problem iteratively, ensuring feasibility and convergence.
• Incorporate a closed-form solution for optimal CPU frequency derived from complementary slackness and KKT conditions, enabling efficient computation.
• Apply a sensitivity analysis to evaluate the impact of image acquisition parameters on energy efficiency and system scalability.

Experimental results

Research questions

• RQ1 How can inter-satellite links be leveraged to reduce downlink congestion in high-resolution Earth observation missions?
• RQ2 What is the optimal trade-off between image compression ratio, processing frequency, and energy consumption in a distributed SMEC framework?
• RQ3 To what extent can distributed SMEC outperform direct downlink and local SMEC in terms of image throughput and energy efficiency?
• RQ4 How does joint optimization of data distribution, compression, and processing frequency affect system performance compared to independent frame optimization?
• RQ5 What energy savings are achievable in real-world scenarios, such as monitoring a volcanic island, through intelligent resource allocation?

Key findings

• The proposed SMEC framework increases the number of images the system can support by 12× compared to direct downlink transmission.
• The framework achieves a 2× improvement in image throughput over local SMEC, where processing is done only on individual satellites.
• In a real-life scenario of imaging a volcanic island, energy consumption was reduced by 11% through joint optimization of task parameters.
• A sensitivity analysis demonstrated that energy consumption can be reduced by up to 90% by carefully selecting data allocation, compression ratios, and processing frequencies.
• The optimal CPU frequency is uniformly distributed across frames within each satellite task, and the closed-form solution ensures computational efficiency.
• Joint optimization of all parameters leads to significant energy savings, and optimizing each frame independently fails to achieve the same efficiency gains.