Deep Joint Source-Channel Coding for Small Satellite Applications

2025-08-01

AI Summary

To address the high-dimensional remote sensing data transmission bottleneck in low-Earth-orbit (LEO) small satellites—caused by short pass durations and highly time-varying channels—this paper proposes ADJSCC-SAT, an adaptive deep joint source-channel coding framework tailored for dynamic channel environments. Built upon an end-to-end trainable neural network with integrated attention mechanisms, ADJSCC-SAT enables a single model to adapt in real time to diverse channel states (e.g., varying SNR levels and fading types), significantly enhancing robustness against channel estimation errors while reducing storage overhead. Experiments on Sentinel-2 multispectral imagery demonstrate that ADJSCC-SAT achieves reconstruction quality comparable to dedicated multi-model systems, with an average PSNR gain of 1.2 dB and a 76% reduction in model parameters. This work presents the first lightweight, highly adaptive deployment of deep joint source-channel coding under realistic satellite-to-ground link constraints, establishing a new paradigm for onboard intelligent compression and real-time transmission.

Technology Category

• Computer Vision: Remote Sensing / Geospatial AI
• Machine Learning: Learning on the Edge & Model Compression
• Cognitive Modeling & Cognitive Systems: Neural Spike Coding

Application Category

• Systems and Infrastructure for Web, Mobile and WoT: Applied ML and AI for Web-based mobile applications
• Search and Retrieval-Augmented AI: Efficiency and scalability of Web search engines
• Economics, Online Markets and Human Computation: Cost models of using LLMs in production systems

Abstract

Small satellites used for Earth observation generate vast amounts of high-dimensional data, but their operation in low Earth orbit creates a significant communication bottleneck due to limited contact times and harsh, varying channel conditions. While deep joint source-channel coding (DJSCC) has emerged as a promising technique, its practical application to the complex satellite environment remains an open question. This paper presents a comprehensive DJSCC framework tailored for satellite communications. We first establish a basic system, DJSCC-SAT, and integrate a realistic, multi-state statistical channel model to guide its training and evaluation. To overcome the impracticality of using separate models for every channel condition, we then introduce an adaptable architecture, ADJSCC-SAT, which leverages attention modules to allow a single neural network to adjust to a wide range of channel states with minimal overhead. Through extensive evaluation on Sentinel-2 multi-spectral data, we demonstrate that our adaptable approach achieves performance comparable to using multiple specialized networks while significantly reducing model storage requirements. Furthermore, the adaptable model shows enhanced robustness to channel estimation errors, outperforming the non-adaptable baseline. The proposed framework is a practical and efficient step toward deploying robust, adaptive DJSCC systems for real-world satellite missions.

Problem

• Addressing communication bottlenecks in small satellite data transmission
• Adapting deep joint source-channel coding to satellite environments
• Reducing model storage while maintaining performance in varying channels

Innovation

• Tailored DJSCC framework for satellite communications
• Adaptable architecture with attention modules
• Enhanced robustness to channel estimation errors