# Massive MIMO Channel Estimation with an Untrained Deep Neural Network

arXiv: https://arxiv.org/abs/1908.00144 · HTML: https://mlsanity.com/paper/1908.00144/

## Key facts

- Cited by 21 papers in the map (data through 14 Oct 2026).
- First posted on arXiv 31 Jul 2019.

## Built on by (most cited first, 21 of 21)

- [Leveraging triplet loss and nonlinear dimensionality reduction for on-the-fly channel charting](https://mlsanity.com/paper/2204.13996/) (2022)
- [Channel charting based beamforming](https://mlsanity.com/paper/2212.12340/) (2022)
- [Physically Parameterized Differentiable MUSIC for DoA Estimation with Uncalibrated Arrays](https://mlsanity.com/paper/2411.15144/) (2024)
- [Achieving Robust Generalization for Wireless Channel Estimation Neural Networks by Designed Training Data](https://mlsanity.com/paper/2302.02302/) (2023)
- [Model-based learning for location-to-channel mapping](https://mlsanity.com/paper/2308.14370/) (2023)
- [Model-based Deep Learning for Beam Prediction based on a Channel Chart](https://mlsanity.com/paper/2312.02239/) (2023)
- [Model-based learning for multi-antenna multi-frequency location-to-channel mapping](https://mlsanity.com/paper/2407.07719/) (2024)
- [Deep Learning-based Techniques for Integrated Sensing and Communication Systems: State-of-the-Art, Challenges, and Opportunities](https://mlsanity.com/paper/2509.06968/) (2025)
- [End-to-End Learning for Uplink MU-SIMO Joint Transmitter and Non-Coherent Receiver Design in Fading Channels](https://mlsanity.com/paper/2105.01260/) (2021)
- [Nested Construction of Polar Codes via Transformers](https://mlsanity.com/paper/2401.17188/) (2024)
- [Untrained DNN for Channel Estimation of RIS-Assisted Multi-User OFDM System with Hardware Impairments](https://mlsanity.com/paper/2107.07423/) (2021)
- [Transfer Learning Capabilities of Untrained Neural Networks for MIMO CSI Recreation](https://mlsanity.com/paper/2111.07858/) (2021)
- [Deep Learning of Transferable MIMO Channel Modes for 6G V2X Communications](https://mlsanity.com/paper/2108.13831/) (2021)
- [Machine Learning for CSI Recreation Based on Prior Knowledge](https://mlsanity.com/paper/2111.07854/) (2021)
- [Learning to Precode for Integrated Sensing and Communications Systems](https://mlsanity.com/paper/2303.06381/) (2023)
- [Machine Learning-based Methods for Joint {Detection-Channel Estimation} in OFDM Systems](https://mlsanity.com/paper/2304.12189/) (2023)
- [Physics-Inspired Deep Learning Anti-Aliasing Framework in Efficient Channel State Feedback](https://mlsanity.com/paper/2403.08133/) (2024)
- [A Machine Learning Approach for Simultaneous Demapping of QAM and APSK Constellations](https://mlsanity.com/paper/2405.09909/) (2024)
- [ISDNN: A Deep Neural Network for Channel Estimation in Massive MIMO systems](https://mlsanity.com/paper/2410.20110/) (2024)
- [A CNN-based End-to-End Learning for RIS-assisted Communication System](https://mlsanity.com/paper/2503.13976/) (2025)
- [DiffPace: Diffusion-based Plug-and-play Augmented Channel Estimation in mmWave and Terahertz Ultra-Massive MIMO Systems](https://mlsanity.com/paper/2511.01867/) (2025)

## Abstract (arXiv)

This paper proposes a deep learning-based channel estimation method for multi-cell interference-limited massive MIMO systems, in which base stations equipped with a large number of antennas serve multiple single-antenna users. The proposed estimator employs a specially designed deep neural network (DNN) to first denoise the received signal, followed by a conventional least-squares (LS) estimation. We analytically prove that our LS-type deep channel estimator can approach minimum mean square error (MMSE) estimator performance for high-dimensional signals, while avoiding MMSE's requirement for complex channel inversions and knowledge of the channel covariance matrix. This analytical result, while asymptotic, is observed in simulations to be operational for just 64 antennas and 64 subcarriers per OFDM symbol. The proposed method also does not require any training and utilizes several orders of magnitude fewer parameters than conventional DNNs. The proposed deep channel estimator is also robust to pilot contamination and can even completely eliminate it under certain conditions.