FOOL: Addressing the Downlink Bottleneck in Satellite Computing with Neural Feature Compression

2024-03-25

IEEE Transactions on Mobile Computing

AI Summary

To address the downlink bandwidth bottleneck in low-Earth-orbit (LEO) nanosatellite constellations, existing orbital edge computing (OEC) approaches suffer from limited generalizability and practicality due to reliance on task-specific priors or coarse-grained filtering. This paper proposes FOOL: a native OEC, task-agnostic, onboard neural feature compression method. Its key innovations include—firstly, explicitly modeling LEO link intermittency within the compression framework; secondly, integrating block-wise encoding, context-aware embedding, and cross-block dependency modeling to achieve high-fidelity feature compression and low-bitrate reversible reconstruction under stringent nanosatellite hardware constraints. Experiments demonstrate that FOOL achieves over 100× reduction in downlink data volume without requiring downstream task priors, attains reconstruction quality (PSNR/SSIM) competitive with state-of-the-art methods, significantly reduces bitrate, and has been validated on a standard nanosatellite platform.

Technology Category

• Machine Learning: Learning on the Edge & Model Compression
• Data Mining & Knowledge Management: Data Compression
• Search and Optimization: Local Search

Application Category

• Systems and Infrastructure for Web, Mobile and WoT: Data management and stream processing for Web, mobile and wireless applications
• Economics, Online Markets and Human Computation: Architectures and workflows that use LLMs for crowd work
• Graph Algorithms and Modeling for the Web: Efficient manipulation of static and dynamic Web-related graphs

Abstract

Nanosatellite constellations equipped with sensors capturing large geographic regions provide unprecedented opportunities for Earth observation. As constellation sizes increase, network contention poses a downlink bottleneck. Orbital Edge Computing (OEC) leverages limited onboard compute resources to reduce transfer costs by processing the raw captures at the source. However, current solutions have limited practicability due to reliance on crude filtering methods or over-prioritizing particular downstream tasks. This work presents FOOL, an OEC-native and task-agnostic feature compression method that preserves prediction performance. FOOL partitions high-resolution satellite imagery to maximize throughput. Further, it embeds context and leverages inter-tile dependencies to lower transfer costs with negligible overhead. While FOOL is a feature compressor, it can recover images with competitive scores on quality measures at lower bitrates. We extensively evaluate transfer cost reduction by including the peculiarity of intermittently available network connections in low earth orbit. Lastly, we test the feasibility of our system for standardized nanosatellite form factors. We demonstrate that FOOL permits downlinking over 100x the data volume without relying on prior information on the downstream tasks.

Problem

• Addressing downlink bottleneck in satellite computing
• Reducing transfer costs with neural feature compression
• Preserving prediction performance without task prioritization

Innovation

• Neural feature compression for satellite imagery
• Task-agnostic OEC-native method preserves prediction performance
• Partitions imagery and leverages inter-tile dependencies