DynaFilter: Cloud-driven Dynamic Filtering for Satellite Edge Intelligence

2026-07-10

AI Summary

This work addresses the inefficiencies of existing remote sensing data transmission methods in satellite edge systems, which are constrained by limited bandwidth and intermittent connectivity—either requiring full decompression for preprocessing or uploading compressed data indiscriminately, leading to significant resource waste. The paper proposes a novel dynamic semantic filtering mechanism operating directly in the compressed domain, establishing, for the first time, a precise mapping between high-level semantic queries from the cloud and multimodal compressed-domain features (e.g., JPEG DC/AC coefficients and video motion vectors). By integrating lightweight feature extraction, compressed-domain region selection, and cloud-edge semantic alignment, the approach enables direct region-of-interest inference without decompression, reducing decoded pixel volume by 1.6–7.1× for image tasks, cutting video bandwidth usage by 92.0%, lowering energy consumption by 43.1–88.6%, and accelerating inference latency by 1.6–3.0×.

Abstract

Modern satellite edge systems, including those performing remote sensing tasks such object detection and tracking, are characterized by severely limited bandwidth and intermittent connections, making continuous data transmission to the cloud impractical. Existing edge-cloud systems, however, either require heavy pre-processing before analysis, for instance, full decompression of imagery data, or transmit all compressed data regardless of relevance. To address these challenges, we design DynaFilter, a dynamic filtering technique that enables satellite edge devices to perform selective region-of-interest (RoI) inference directly in the compressed-domain, without full decompression. Our key insight is that low-level compression syntax, specifically DC coefficients/AC energy in JPEG images and motion vectors in video streams, exhibits strong correlations with high-level semantic queries. By establishing a precise mapping between cloud query semantics and multimodal compressed-domain features, DynaFilter enables the edge to identify and transmit only relevant data associated to RoIs. Extensive evaluations show that DynaFilter reduces the total volume of pixel data for decoding and subsequent inference by 1.6x-7.1x for images, and achieves 92.0% bandwidth savings for video streams compared to state-of-the-art baselines. Furthermore, it decreases energy consumption by 43.1-88.6% on target devices and achieves a 1.6x-3.0x speedup in inference latency.

Problem

• satellite edge intelligence
• dynamic filtering
• compressed-domain processing
• bandwidth limitation
• region-of-interest

Innovation

• compressed-domain inference
• region-of-interest filtering
• satellite edge intelligence
• dynamic filtering
• cloud-edge collaboration