Generative Decoding of Compressed CSI for MIMO Precoding Design

Fuente: arXiv
Saved in:
Bibliographic Details
Main Authors: Luo, Hao, Khosravirad, Saeed R., Alkhateeb, Ahmed
Format: Preprint
Published: 2025
Subjects:
Online Access:
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866911259026259968
author Luo, Hao
Khosravirad, Saeed R.
Alkhateeb, Ahmed
author_facet Luo, Hao
Khosravirad, Saeed R.
Alkhateeb, Ahmed
contents Massive MIMO systems can enhance spectral and energy efficiency, but they require accurate channel state information (CSI), which becomes costly as the number of antennas increases. While machine learning (ML) autoencoders show promise for CSI reconstruction and reducing feedback overhead, they introduce new challenges with standardization, interoperability, and backward compatibility. Also, the significant data collection needed for training makes real-world deployment difficult. To overcome these drawbacks, we propose an ML-based, decoder-only solution for compressed CSI. Our approach uses a standardized encoder for CSI compression on the user side and a site-specific generative decoder at the base station to refine the compressed CSI using environmental knowledge. We introduce two training schemes for the generative decoder: An end-to-end method and a two-stage method, both utilizing a goal-oriented loss function. Furthermore, we reduce the data collection overhead by using a site-specific digital twin to generate synthetic CSI data for training. Our simulations highlight the effectiveness of this solution across various feedback overhead regimes.
format Preprint
id arxiv_https___arxiv_org_abs_2511_07783
institution arXiv
publishDate 2025
record_format arxiv
spellingShingle Generative Decoding of Compressed CSI for MIMO Precoding Design
Luo, Hao
Khosravirad, Saeed R.
Alkhateeb, Ahmed
Signal Processing
Information Theory
Massive MIMO systems can enhance spectral and energy efficiency, but they require accurate channel state information (CSI), which becomes costly as the number of antennas increases. While machine learning (ML) autoencoders show promise for CSI reconstruction and reducing feedback overhead, they introduce new challenges with standardization, interoperability, and backward compatibility. Also, the significant data collection needed for training makes real-world deployment difficult. To overcome these drawbacks, we propose an ML-based, decoder-only solution for compressed CSI. Our approach uses a standardized encoder for CSI compression on the user side and a site-specific generative decoder at the base station to refine the compressed CSI using environmental knowledge. We introduce two training schemes for the generative decoder: An end-to-end method and a two-stage method, both utilizing a goal-oriented loss function. Furthermore, we reduce the data collection overhead by using a site-specific digital twin to generate synthetic CSI data for training. Our simulations highlight the effectiveness of this solution across various feedback overhead regimes.
title Generative Decoding of Compressed CSI for MIMO Precoding Design
topic Signal Processing
Information Theory
url https://arxiv.org/abs/2511.07783