# Study on Downlink CSI compression: Are Neural Networks the Only Solution?

arXiv: https://arxiv.org/abs/2502.17459 · HTML: https://mlsanity.com/paper/2502.17459/

## Key facts

- No citing papers in the map yet (data through 14 Oct 2026).
- Builds on 7 of its 18 references that are in the map.
- First posted on arXiv 10 Feb 2025.

## Builds on (most cited first, 7 of 7)

- [Deep Residual Learning for Image Recognition](https://mlsanity.com/paper/1512.03385/) (2015)
- [Distributed Deep Convolutional Compression for Massive MIMO CSI Feedback](https://mlsanity.com/paper/2003.04684/) (2020)
- [Deep Convolutional Compression for Massive MIMO CSI Feedback](https://mlsanity.com/paper/1907.02942/) (2019)
- [Deep CSI Compression for Massive MIMO: A Self-information Model-driven Neural Network](https://mlsanity.com/paper/2204.11567/) (2022)
- [AnciNet: An Efficient Deep Learning Approach for Feedback Compression of Estimated CSI in Massive MIMO Systems](https://mlsanity.com/paper/2008.07112/) (2020)
- [Learning-Based Phase Compression and Quantization for Massive MIMO CSI Feedback with Magnitude-Aided Information](https://mlsanity.com/paper/2103.00432/) (2021)
- [Dealing with CSI Compression to Reduce Losses and Overhead: An Artificial Intelligence Approach](https://mlsanity.com/paper/2104.00189/) (2021)

## Abstract (arXiv)

Massive Multi Input Multi Output (MIMO) systems enable higher data rates in the downlink (DL) with spatial multiplexing achieved by forming narrow beams. The higher DL data rates are achieved by effective implementation of spatial multiplexing and beamforming which is subject to availability of DL channel state information (CSI) at the base station. For Frequency Division Duplexing (FDD) systems, the DL CSI has to be transmitted by User Equipment (UE) to the gNB and it constitutes a significant overhead which scales with the number of transmitter antennas and the granularity of the CSI. To address the overhead issue, AI/ML methods using auto-encoders have been investigated, where an encoder neural network model at the UE compresses the CSI and a decoder neural network model at the gNB reconstructs it. However, the use of AI/ML methods has a number of challenges related to (1) model complexity, (2) model generalization across channel scenarios and (3) inter-vendor compatibility of the two sides of the model. In this work, we investigate a more traditional dimensionality reduction method that uses Principal Component Analysis (PCA) and therefore does not suffer from the above challenges. Simulation results show that PCA based CSI compression actually achieves comparable reconstruction performance to commonly used deep neural networks based models.