mlsanityData to 14 Oct 2026

Denoising Refinement Diffusion Models for Simultaneous Generation of Multi-scale Mobile Network Traffic

arXiv 2511.17532 ↗First posted 30 Oct 2025cs.NI cs.AIcited by 0

On the map: Diffusion models → Diffusion and flow policies for robotics

Abstract (arXiv)

The planning, management, and resource scheduling of cellular mobile networks require joint estimation of mobile traffic across different layers and nodes. Mobile traffic generation can proactively anticipate user demands and capture the dynamics of network load. However, existing methods mainly focus on generating traffic at a single spatiotemporal resolution, making it difficult to jointly model multi-scale traffic patterns. In this paper, we propose ZoomDiff, a diffusion-based model for multi-scale mobile traffic generation. ZoomDiff maps urban environmental context into mobile traffic with multiple spatial and temporal resolutions through a set of customized Denoising Refinement Diffusion Models (DRDM). DRDM employs a multi-stage noise-adding and denoising mechanism, enabling different stages to generate traffic at distinct spatiotemporal resolutions. This design aligns the progressive denoising process with hierarchical network layers, including base stations, cells, and grids of varying granularities. Experiments on real-world mobile traffic datasets show that ZoomDiff achieves at least an 18.4% improvement over state-of-the-art baselines in multi-scale traffic generation tasks. Moreover, ZoomDiff demonstrates strong efficiency and cross-city generalization, highlighting its potential as a powerful generative framework for modeling multi-scale mobile network dynamics.

Key facts. No citing papers in the map yet (data through 14 Oct 2026). Placed in Diffusion and flow policies for robotics, one of 14 directions in Diffusion models. Builds on 32 of its 78 references that are in the map. First posted on arXiv 30 Oct 2025.

Builds on

32 of 78 references.

Closest ancestors · 2

Same direction, newest first

  1. 2025Unveiling the Power of Noise Priors: Enhancing Diffusion Models for Mobile Traffic Prediction23 Jan
  2. 2023NetDiffus: Network Traffic Generation by Diffusion Models through Time-Series Imaging23 Sep

Other references · 30

  1. 2017Attention Is All You Needcited by 60,361
  2. 2020Denoising Diffusion Probabilistic Modelscited by 21,076
  3. 2022Scalable Diffusion Models with Transformerscited by 6,054
  4. 2016XGBoost: A Scalable Tree Boosting Systemcited by 5,166
  5. 2020Informer: Beyond Efficient Transformer for Long Sequence Time-Series Forecastingcited by 2,169
  6. 2023iTransformer: Inverted Transformers Are Effective for Time Series Forecastingcited by 931
  7. 2021CSDI: Conditional Score-based Diffusion Models for Probabilistic Time Series Imputationcited by 582
  8. 2016Deep Spatio-Temporal Residual Networks for Citywide Crowd Flows Predictioncited by 333
  9. 2024Diffusion-TS: Interpretable Diffusion for General Time Series Generationcited by 184
  10. 2023PDFormer: Propagation Delay-Aware Dynamic Long-Range Transformer for Traffic Flow Predictioncited by 137
  11. 2022Scaleformer: Iterative Multi-scale Refining Transformers for Time Series Forecastingcited by 54
  12. 2023Spatio-temporal Diffusion Point Processescited by 52
18 more
  1. 2020Deep Analysis of CNN-based Spatio-temporal Representations for Action Recognitioncited by 45
  2. 2022Deep Latent State Space Models for Time-Series Generationcited by 38
  3. 2024MG-TSD: Multi-Granularity Time Series Diffusion Models with Guided Learning Processcited by 32
  4. 2023NetDiffusion: Network Data Augmentation Through Protocol-Constrained Traffic Generationcited by 27
  5. 2018Attentive Crowd Flow Machinescited by 22
  6. 2023Towards Generative Modeling of Urban Flow through Knowledge-enhanced Denoising Diffusioncited by 17
  7. 2021RNN with Particle Flow for Probabilistic Spatio-temporal Forecastingcited by 16
  8. 2024BIGCity: A Universal Spatiotemporal Model for Unified Trajectory and Traffic State Data Analysiscited by 16
  9. 2020STAN: Synthetic Network Traffic Generation with Generative Neural Modelscited by 11
  10. 2019Approximate Query Processing using Deep Generative Modelscited by 8
  11. 2019D-GAN: Deep Generative Adversarial Nets for Spatio-Temporal Predictioncited by 8
  12. 2022Joint Base Station and IRS Deployment for Enhancing Network Coverage: A Graph-Based Modeling and Optimization Approachcited by 5
  13. 2024Generative Models for Synthetic Urban Mobility Data: A Systematic Literature Reviewcited by 5
  14. 2022Efficient Content Delivery in User-Centric and Cache-Enabled Vehicular Edge Networks with Deadline-Constrained Heterogeneous Demandscited by 3
  15. 2025Predicting the Dynamics of Complex System via Multiscale Diffusion Autoencodercited by 3
  16. 2020Faster and More Accurate Measurement through Additive-Error Counterscited by 2
  17. 2021User Localization Based on Call Detail Recordscited by 1
  18. 2023Joint Optimization of Resource Allocation and User Association in Multi-Frequency Cellular Networks Assisted by RIScited by 1

Built on by

Not cited yet.

Similar, not cited

Close in wording, no citation either way

  1. 2025Multi-Scale Diffusion Transformer for Jointly Simulating User Mobility and Mobile Traffic Pattern
  2. 2026RadioDiff-Flux: Efficient Radio Map Construction via Generative Denoise Diffusion Model Trajectory Midpoint Reuse
  3. 2024Diffusion Models Meet Network Management: Improving Traffic Matrix Analysis with Diffusion-based Approach
  4. 2025LSDM: LLM-Enhanced Spatio-temporal Diffusion Model for Service-Level Mobile Traffic Prediction
  5. 2024Noise Matters: Diffusion Model-based Urban Mobility Generation with Collaborative Noise Priors
  6. 2023Complexity-aware Large Scale Origin-Destination Network Generation via Diffusion Model
  7. 2026A Multi-Resolution Multi-Domain Pre-Training Framework for Universal Traffic Forecasting
  8. 2026RMPrior: Bridging Propagation Priors and Diffusion Refinement for Efficient Radio Map Construction