mlsanityData to 7 Oct 2026

Over-the-Air Design of GAN Training for mmWave MIMO Channel Estimation

arXiv 2205.12445 ↗First posted 25 May 2022eess.SP cs.IT cs.LGcited by 5

Not on the map. Why?

Abstract (arXiv)

Future wireless systems are trending towards higher carrier frequencies that offer larger communication bandwidth but necessitate the use of large antenna arrays. Existing signal processing techniques for channel estimation do not scale well to this "high-dimensional" regime in terms of performance and pilot overhead. Meanwhile, training deep learning based approaches for channel estimation requires large labeled datasets mapping pilot measurements to clean channel realizations, which can only be generated offline using simulated channels. In this paper, we develop a novel unsupervised over-the-air (OTA) algorithm that utilizes noisy received pilot measurements to train a deep generative model to output beamspace MIMO channel realizations. Our approach leverages Generative Adversarial Networks (GAN), while using a conditional input to distinguish between Line-of-Sight (LOS) and Non-Line-of-Sight (NLOS) channel realizations. We also present a federated implementation of the OTA algorithm that distributes the GAN training over multiple users and greatly reduces the user side computation. We then formulate channel estimation from a limited number of pilot measurements as an inverse problem and reconstruct the channel by optimizing the input vector of the trained generative model. Our proposed approach significantly outperforms Orthogonal Matching Pursuit on both LOS and NLOS channel models, and EM-GM-AMP -- an Approximate Message Passing algorithm -- on LOS channel models, while achieving comparable performance on NLOS channel models in terms of the normalized channel reconstruction error. More importantly, our proposed framework has the potential to be trained online using real noisy pilot measurements, is not restricted to a specific channel model and can even be utilized for a federated OTA design of a dataset generator from noisy data.

Key facts. Cited by 5 papers in the map (data through 7 Oct 2026). Builds on 32 of its 61 references that are in the map. First posted on arXiv 25 May 2022.

Builds on

32 of 61 references.

References · 32

  1. 2014Adam: A Method for Stochastic Optimizationcited by 36,960
  2. 2019PyTorch: An Imperative Style, High-Performance Deep Learning Librarycited by 20,160
  3. 2015Batch Normalization: Accelerating Deep Network Training by Reducing Internal Covariate Shiftcited by 6,979
  4. 2022Generative Adversarial Networkscited by 5,603
  5. 2017Improved Training of Wasserstein GANscited by 2,223
  6. 2014Conditional Generative Adversarial Netscited by 2,152
  7. 2015Unsupervised Representation Learning with Deep Convolutional Generative Adversarial Networkscited by 2,048
  8. 2019Towards Federated Learning at Scale: System Designcited by 826
  9. 2017Compressed Sensing using Generative Modelscited by 337
  10. 2017Real-valued (Medical) Time Series Generation with Recurrent Conditional GANscited by 329
  11. 2020Deep Learning Techniques for Inverse Problems in Imagingcited by 290
  12. 2021Robust Compressed Sensing MRI with Deep Generative Priorscited by 284
20 more
  1. 2019Model Pruning Enables Efficient Federated Learning on Edge Devicescited by 219
  2. 2017VEEGAN: Reducing Mode Collapse in GANs using Implicit Variational Learningcited by 199
  3. 2019Generative Models for Effective ML on Private, Decentralized Datasetscited by 85
  4. 2014Convolutional Neural Networks at Constrained Time Costcited by 81
  5. 2021Adaptive Quantization of Model Updates for Communication-Efficient Federated Learningcited by 60
  6. 2018MD-GAN: Multi-Discriminator Generative Adversarial Networks for Distributed Datasetscited by 49
  7. 2020Robust Compressed Sensing using Generative Modelscited by 33
  8. 2020High Dimensional Channel Estimation Using Deep Generative Networkscited by 33
  9. 2018Task-Aware Compressed Sensing with Generative Adversarial Networkscited by 17
  10. 2020Wideband Channel Estimation with A Generative Adversarial Networkcited by 16
  11. 2018MEGAN: Mixture of Experts of Generative Adversarial Networks for Multimodal Image Generationcited by 15
  12. 2018Physical Layer Communications System Design Over-the-Air Using Adversarial Networkscited by 12
  13. 2020WGAN-based Autoencoder Training Over-the-aircited by 11
  14. 2022MIMO-GAN: Generative MIMO Channel Modelingcited by 9
  15. 2016Initial Access in 5G mm-Wave Cellular Networkscited by 8
  16. 2019Robust One-Bit Recovery via ReLU Generative Networks: Near-Optimal Statistical Rate and Global Landscape Analysiscited by 8
  17. 2021Score-Based Generative Models for Robust Channel Estimationcited by 8
  18. 2020DeepWiPHY: Deep Learning-based Receiver Design and Dataset for IEEE 802.11ax Systemscited by 6
  19. 2019Low-rank mmWave MIMO channel estimation in one-bit receiverscited by 1
  20. 2019Model-Driven Channel Estimation for OFDM Systems Based on Image Super-Resolution Networkcited by 1

Built on by

Cited by year

Citing papers per year: 2023: 1; 2024: 2; 2025: 2.

5 papers.

  1. 2024Generative Diffusion Models for High Dimensional Channel Estimationcited by 26
  2. 2024Generating High Dimensional User-Specific Wireless Channels using Diffusion Modelscited by 14
  3. 2023Leveraging Variational Autoencoders for Parameterized MMSE Estimationcited by 6
  4. 2025A Comprehensive Survey on Network Traffic Synthesis: From Statistical Models to Deep Learningcited by 2
  5. 2025On optimal solutions of classical and sliced Wasserstein GANs with non-Gaussian datacited by 0