mlsanityData to 14 Oct 2026

Diffusion Models for Accurate Channel Distribution Generation

arXiv 2309.10505 ↗First posted 19 Sep 2023cs.IT cs.LG math.ITcited by 5

On the map: Diffusion models → Diffusion for compression and wireless

Abstract (arXiv)

Strong generative models can accurately learn channel distributions. This could save recurring costs for physical measurements of the channel. Moreover, the resulting differentiable channel model supports training neural encoders by enabling gradient-based optimization. The initial approach in the literature draws upon the modern advancements in image generation, utilizing generative adversarial networks (GANs) or their enhanced variants to generate channel distributions. In this paper, we address this channel approximation challenge with diffusion models (DMs), which have demonstrated high sample quality and mode coverage in image generation. In addition to testing the generative performance of the channel distributions, we use an end-to-end (E2E) coded-modulation framework underpinned by DMs and propose an efficient training algorithm. Our simulations with various channel models show that a DM can accurately learn channel distributions, enabling an E2E framework to achieve near-optimal symbol error rates (SERs). Furthermore, we examine the trade-off between mode coverage and sampling speed through skipped sampling using sliced Wasserstein distance (SWD) and the E2E SER. We investigate the effect of noise scheduling on this trade-off, demonstrating that with an appropriate choice of parameters and techniques, sampling time can be significantly reduced with a minor increase in SWD and SER. Finally, we show that the DM can generate a correlated fading channel, whereas a strong GAN variant fails to learn the covariance. This paper highlights the potential benefits of using DMs for learning channel distributions, which could be further investigated for various channels and advanced techniques of DMs.

Key facts. Cited by 5 papers in the map (data through 14 Oct 2026). Placed in Diffusion for compression and wireless, one of 14 directions in Diffusion models. Builds on 33 of its 63 references that are in the map. First posted on arXiv 19 Sep 2023.

Builds on

33 of 63 references.

Closest ancestors · 4

Same direction, newest first

  1. 2023CDDM: Channel Denoising Diffusion Models for Wireless Semantic Communications16 Sep
  2. 2023Adversarial Score-Based Generative Models for MMSE-achieving AmBC Channel Estimation11 Sep
  3. 2023Generative Diffusion Models for Radio Wireless Channel Modelling and Sampling10 Aug
  4. 2022MIMO Channel Estimation using Score-Based Generative Models14 Apr

Other references · 29

  1. 2014Adam: A Method for Stochastic Optimizationcited by 37,240
  2. 2020Denoising Diffusion Probabilistic Modelscited by 21,076
  3. 2015U-Net: Convolutional Networks for Biomedical Image Segmentationcited by 19,713
  4. 2020Denoising Diffusion Implicit Modelscited by 9,781
  5. 2020Score-Based Generative Modeling through Stochastic Differential Equationscited by 9,122
  6. 2015Deep Unsupervised Learning using Nonequilibrium Thermodynamicscited by 7,058
  7. 2019Generative Modeling by Estimating Gradients of the Data Distributioncited by 4,101
  8. 2021Improved Denoising Diffusion Probabilistic Modelscited by 3,761
  9. 2017FiLM: Visual Reasoning with a General Conditioning Layercited by 2,333
  10. 2015Unsupervised Representation Learning with Deep Convolutional Generative Adversarial Networkscited by 2,054
  11. 2021Variational Diffusion Modelscited by 1,153
  12. 2022Diffusion Models in Vision: A Surveycited by 1,015
17 more
  1. 2023Common Diffusion Noise Schedules and Sample Steps are Flawedcited by 310
  2. 2017An Introduction to Deep Learning for the Physical Layercited by 233
  3. 2016Generating Images with Perceptual Similarity Metrics based on Deep Networkscited by 222
  4. 2019Turbo Autoencoder: Deep learning based channel codes for point-to-point communication channelscited by 44
  5. 2020High Dimensional Channel Estimation Using Deep Generative Networkscited by 35
  6. 2018Approximating the Void: Learning Stochastic Channel Models from Observation with Variational Generative Adversarial Networkscited by 25
  7. 2020Wideband Channel Estimation with A Generative Adversarial Networkcited by 16
  8. 2018Physical Layer Communications System Design Over-the-Air Using Adversarial Networkscited by 12
  9. 2020WGAN-based Autoencoder Training Over-the-aircited by 11
  10. 2020Generative Neural Network Channel Modeling for Millimeter-Wave UAV Communicationcited by 11
  11. 2022MIMO-GAN: Generative MIMO Channel Modelingcited by 9
  12. 2020Deep Energy Autoencoder for Noncoherent Multicarrier MU-SIMO Systemscited by 7
  13. 2019Autoencoder-Based Error Correction Coding for One-Bit Quantizationcited by 6
  14. 2018Achievable Information Rates for Nonlinear Fiber Communication via End-to-end Autoencoder Learningcited by 5
  15. 2018Performance Analysis of Deep Learning based on Recurrent Neural Networks for Channel Codingcited by 4
  16. 2022Hybrid Neural Coded Modulation: Design and Training Methodscited by 2
  17. 2023ProductAE: Toward Deep Learning Driven Error-Correction Codes of Large Dimensionscited by 1

Built on by

Cited by year (2026 to 14 Oct)

Citing papers per year: 2023: 1; 2024: 1; 2025: 1; 2026: 2.

5 papers.

  1. 2024Diffusion-based Generative Prior for Low-Complexity MIMO Channel Estimationcited by 25
  2. 2025Generative Diffusion Models for Wireless Networks: Fundamental, Architecture, and State-of-the-Artcited by 14
  3. 2023Generative AI-Based Probabilistic Constellation Shaping With Diffusion Modelscited by 2
  4. 2026EVT-Based Generative AI for Tail-Aware Channel Estimationcited by 1
  5. 2026Condition-Wise Sinkhorn Drifting for One-Shot Learned Channel Simulationcited by 0

Similar, not cited

Close in wording, no citation either way

  1. 2023Learning End-to-End Channel Coding with Diffusion Models
  2. 2024Generating High Dimensional User-Specific Wireless Channels using Diffusion Models
  3. 2024You Only Sample Once: Taming One-Step Text-to-Image Synthesis by Self-Cooperative Diffusion GANs
  4. 2024Generative Diffusion Models for High Dimensional Channel Estimation
  5. 2024Adv-KD: Adversarial Knowledge Distillation for Faster Diffusion Sampling
  6. 2023Efficient Transfer Learning in Diffusion Models via Adversarial Noise
  7. 2025Adaptive Destruction Processes for Diffusion Samplers
  8. 2025Revisiting Diffusion Models: From Generative Pre-training to One-Step Generation