Variable-Rate DDM Compression Using Modulated Autoencoder with Window-Based Attention

View at publisher Summarise with VIRAT
Authors: Chun-Han Chen, Pei‐Jun Lee
Cited by: 0
References: 6

Summary

No summary yet. VIRAT can read this paper and write a plain-language summary of its aim, methods, key findings and limitations.

Summarise with VIRAT Free to try with your Science Catalogs account.

Abstract

This paper addresses the bandwidth and storage constraints in GNSS-Reflectometry missions by proposing a modulated autoencoder with window-based attention for Delay-Doppler Map (DDM) compression. The model enables variable-rate compression through a modulation network with single weights and enhances critical reflection regions (glistening and annulus zones) via localized attention. Experimental results show that it achieves superior rate–distortion performance compared with JPEG2000 and WebP. Moreover, wind speed retrieval using reconstructed DDMs incurs negligible accuracy loss, confirming the scientific reliability of the compressed data.

Fields of Study & Subjects

Advanced Data Compression Techniques Video Coding and Compression Technologies Image and Signal Denoising Methods

Publication details

Journal / source
International Conference on Consumer Electronics
Pages
1–2
Publication date
3 February 2026
Language
EN