Generative Diffusion Model-based Compression of MIMO CSI

Fuente: arXiv
Saved in:
Bibliographic Details
Main Authors: Kim, Heasung, Lee, Taekyun, Kim, Hyeji, De Veciana, Gustavo, Arfaoui, Mohamed Amine, Koc, Asil, Pietraski, Phil, Zhang, Guodong, Kaewell, John
Format: Preprint
Published: 2025
Subjects:
Online Access:
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866916644572364800
author Kim, Heasung
Lee, Taekyun
Kim, Hyeji
De Veciana, Gustavo
Arfaoui, Mohamed Amine
Koc, Asil
Pietraski, Phil
Zhang, Guodong
Kaewell, John
author_facet Kim, Heasung
Lee, Taekyun
Kim, Hyeji
De Veciana, Gustavo
Arfaoui, Mohamed Amine
Koc, Asil
Pietraski, Phil
Zhang, Guodong
Kaewell, John
contents While neural lossy compression techniques have markedly advanced the efficiency of Channel State Information (CSI) compression and reconstruction for feedback in MIMO communications, efficient algorithms for more challenging and practical tasks-such as CSI compression for future channel prediction and reconstruction with relevant side information-remain underexplored, often resulting in suboptimal performance when existing methods are extended to these scenarios. To that end, we propose a novel framework for compression with side information, featuring an encoding process with fixed-rate compression using a trainable codebook for codeword quantization, and a decoding procedure modeled as a backward diffusion process conditioned on both the codeword and the side information. Experimental results show that our method significantly outperforms existing CSI compression algorithms, often yielding over twofold performance improvement by achieving comparable distortion at less than half the data rate of competing methods in certain scenarios. These findings underscore the potential of diffusion-based compression for practical deployment in communication systems.
format Preprint
id arxiv_https___arxiv_org_abs_2503_03753
institution arXiv
publishDate 2025
record_format arxiv
spellingShingle Generative Diffusion Model-based Compression of MIMO CSI
Kim, Heasung
Lee, Taekyun
Kim, Hyeji
De Veciana, Gustavo
Arfaoui, Mohamed Amine
Koc, Asil
Pietraski, Phil
Zhang, Guodong
Kaewell, John
Information Theory
Artificial Intelligence
Signal Processing
68P30
I.2.0
While neural lossy compression techniques have markedly advanced the efficiency of Channel State Information (CSI) compression and reconstruction for feedback in MIMO communications, efficient algorithms for more challenging and practical tasks-such as CSI compression for future channel prediction and reconstruction with relevant side information-remain underexplored, often resulting in suboptimal performance when existing methods are extended to these scenarios. To that end, we propose a novel framework for compression with side information, featuring an encoding process with fixed-rate compression using a trainable codebook for codeword quantization, and a decoding procedure modeled as a backward diffusion process conditioned on both the codeword and the side information. Experimental results show that our method significantly outperforms existing CSI compression algorithms, often yielding over twofold performance improvement by achieving comparable distortion at less than half the data rate of competing methods in certain scenarios. These findings underscore the potential of diffusion-based compression for practical deployment in communication systems.
title Generative Diffusion Model-based Compression of MIMO CSI
topic Information Theory
Artificial Intelligence
Signal Processing
68P30
I.2.0
url https://arxiv.org/abs/2503.03753