Learning to Focus Synthetic Aperture Radar On-line with State-Space Models

2026-05-11

AI Summary

This work addresses the high latency and memory overhead of conventional synthetic aperture radar (SAR) focusing methods, which rely on block-based processing and hinder real-time operation in closed-loop cognitive systems. The authors propose the first Online SAR Processor (OSP), modeling SAR imaging as a streaming process and leveraging a lightweight state-space model to generate focused imagery row-by-row during signal acquisition. A compact proxy model is developed via teacher–student knowledge distillation and multi-stage loss training, substantially reducing computational demands. Compared to a row-wise digital signal processing baseline, the proposed approach achieves approximately 70× lower latency and 130× less memory usage, requiring only 16 ms per row and 6 MB of memory on a single AMD CPU core while maintaining image quality sufficient for downstream tasks such as ship detection and flood mapping.

Abstract

Conventional focusing methods for Synthetic Aperture Radar (SAR) employ block processing efficiently but remain latency-heavy processes that prevent the realisation of a closed-loop cognitive SAR vision system. We present the first Online SAR Processor (OSP), an online image-formation framework that treats SAR sensing as a stream and produces focused SAR image output line by line during acquisition. OSP uses a tiny state-space surrogate model trained with teacher-student distillation and multi-stage losses. We evaluate the method on 300GB of SAR data from Maya4, a Sentinel-1-derived dataset containing raw, range-compressed, range-cell-migration-corrected, and azimuth-compressed products. Relative to a linewise digital-signal-processing baseline, OSP delivers approximately 70$\times$ lower latency and 130$\times$ lower memory use; on a single AMD CPU core it processes one row in 16 ms with a memory footprint of 6 MB whilst maintaining a focusing quality high enough to support downstream decisions, which we illustrate with vessel detection and flood-mapping tasks.

Problem

• Synthetic Aperture Radar
• online processing
• latency
• cognitive SAR
• image formation

Innovation

• Online SAR Processing
• State-Space Models
• Teacher-Student Distillation
• Low-Latency Imaging
• Cognitive Radar