mlsanityData to 14 Oct 2026

Time Series Analysis in Frequency Domain: A Survey of Open Challenges, Opportunities and Benchmarks

arXiv 2504.07099 ↗First posted 12 Feb 2025cs.CE cs.LGcited by 10

Not on the map. Why?

Abstract (arXiv)

Frequency-domain analysis has emerged as a powerful paradigm for time series analysis, offering unique advantages over traditional time-domain approaches while introducing new theoretical and practical challenges. This survey provides a comprehensive examination of spectral methods from classical Fourier analysis to modern neural operators, systematically summarizing three open challenges in current research: (1) causal structure preservation during spectral transformations, (2) uncertainty quantification in learned frequency representations, and (3) topology-aware analysis for non-Euclidean data structures. Through rigorous reviewing of over 100 studies, we develop a unified taxonomy that bridges conventional spectral techniques with cutting-edge machine learning approaches, while establishing standardized benchmarks for performance evaluation. Our work identifies key knowledge gaps in the field, particularly in geometric deep learning and quantum-enhanced spectral analysis. The survey offers practitioners a systematic framework for method selection and implementation, while charting promising directions for future research in this rapidly evolving domain.

Key facts. Cited by 10 papers in the map (data through 14 Oct 2026). Builds on 60 of its 182 references that are in the map. First posted on arXiv 12 Feb 2025.

Builds on

60 of 182 references.

References · 60

  1. 2017Attention Is All You Needcited by 60,361
  2. 2020Informer: Beyond Efficient Transformer for Long Sequence Time-Series Forecastingcited by 2,169
  3. 2016Convolutional Neural Networks on Graphs with Fast Localized Spectral Filteringcited by 2,089
  4. 2022A Time Series is Worth 64 Words: Long-term Forecasting with Transformerscited by 2,039
  5. 2021Autoformer: Decomposition Transformers with Auto-Correlation for Long-Term Series Forecastingcited by 1,676
  6. 2020Reformer: The Efficient Transformercited by 1,668
  7. 2018An Empirical Evaluation of Generic Convolutional and Recurrent Networks for Sequence Modelingcited by 1,476
  8. 2021MLP-Mixer: An all-MLP Architecture for Visioncited by 1,446
  9. 2022FEDformer: Frequency Enhanced Decomposed Transformer for Long-term Series Forecastingcited by 1,210
  10. 2017Modeling Long- and Short-Term Temporal Patterns with Deep Neural Networkscited by 879
  11. 2019N-BEATS: Neural basis expansion analysis for interpretable time series forecastingcited by 753
  12. 2022Transformers in Time Series: A Surveycited by 497
38 more
  1. 2020Time Series Forecasting With Deep Learning: A Surveycited by 459
  2. 2021Anomaly Transformer: Time Series Anomaly Detection with Association Discrepancycited by 394
  3. 2019InceptionTime: Finding AlexNet for Time Series Classificationcited by 387
  4. 2021FNet: Mixing Tokens with Fourier Transformscited by 347
  5. 2024TimeMixer: Decomposable Multiscale Mixing for Time Series Forecastingcited by 346
  6. 2021SCINet: Time Series Modeling and Forecasting with Sample Convolution and Interactioncited by 338
  7. 2023Long-term Forecasting with TiDE: Time-series Dense Encodercited by 332
  8. 2024Foundation Models for Time Series Analysis: A Tutorial and Surveycited by 239
  9. 2024Deep Time Series Models: A Comprehensive Survey and Benchmarkcited by 197
  10. 2023FITS: Modeling Time Series with 10k Parameterscited by 195
  11. 2022Less Is More: Fast Multivariate Time Series Forecasting with Light Sampling-oriented MLP Structurescited by 160
  12. 2019Graph Wavelet Neural Networkcited by 131
  13. 2023FourierGNN: Rethinking Multivariate Time Series Forecasting from a Pure Graph Perspectivecited by 127
  14. 2018Multivariate LSTM-FCNs for Time Series Classificationcited by 120
  15. 2024Large Language Models for Time Series: A Surveycited by 114
  16. 2024TSLANet: Rethinking Transformers for Time Series Representation Learningcited by 74
  17. 2018RobustSTL: A Robust Seasonal-Trend Decomposition Algorithm for Long Time Seriescited by 54
  18. 2020From Fourier to Koopman: Spectral Methods for Long-term Time Series Predictioncited by 50
  19. 2022Self-Supervised Time Series Representation Learning via Cross Reconstruction Transformercited by 47
  20. 2024Fredformer: Frequency Debiased Transformer for Time Series Forecastingcited by 45
  21. 2023PastNet: Introducing Physical Inductive Biases for Spatio-temporal Video Predictioncited by 44
  22. 2023A Survey on Deep Learning based Time Series Analysis with Frequency Transformationcited by 40
  23. 2024Adaptive Multi-Scale Decomposition Framework for Time Series Forecastingcited by 34
  24. 2018Forecasting Economics and Financial Time Series: ARIMA vs. LSTMcited by 33
  25. 2021Missing Value Imputation on Multidimensional Time Seriescited by 33
  26. 2022DEPTS: Deep Expansion Learning for Periodic Time Series Forecastingcited by 29
  27. 2024Toward a Better Understanding of Fourier Neural Operators from a Spectral Perspectivecited by 25
  28. 2019STFNets: Learning Sensing Signals from the Time-Frequency Perspective with Short-Time Fourier Neural Networkscited by 23
  29. 2023RecFNO: a resolution-invariant flow and heat field reconstruction method from sparse observations via Fourier neural operatorcited by 19
  30. 2025FreEformer: Frequency Enhanced Transformer for Multivariate Time Series Forecastingcited by 18
  31. 2023Neural Spectral Methods: Self-supervised learning in the spectral domaincited by 14
  32. 2023Spatial-Temporal Graph Learning with Adversarial Contrastive Adaptationcited by 13
  33. 2024Cross-Domain Pre-training with Language Models for Transferable Time Series Representationscited by 12
  34. 2025Efficient Traffic Prediction Through Spatio-Temporal Distillationcited by 12
  35. 2024Spectral-Refiner: Accurate Fine-Tuning of Spatiotemporal Fourier Neural Operator for Turbulent Flowscited by 11
  36. 20220/1 Deep Neural Networks via Block Coordinate Descentcited by 10
  37. 2020STR: Seasonal-Trend Decomposition Using Regressioncited by 9
  38. 2022A time-causal and time-recursive scale-covariant scale-space representation of temporal signals and past timecited by 8

50 most cited of 60 All 60 (JSON)

Built on by

Cited by year (2026 to 14 Oct)

Citing papers per year: 2025: 2; 2026: 8.

10 papers.

  1. 2025FLDmamba: Integrating Fourier and Laplace Transform Decomposition with Mamba for Enhanced Time Series Predictioncited by 4
  2. 2026MODE: Efficient Time Series Prediction with Mamba Enhanced by Low-Rank Neural ODEscited by 2
  3. 2026StaTS: Spectral Trajectory Schedule Learning for Adaptive Time Series Forecasting with Frequency Guided Denoisercited by 1
  4. 2026L-Drive: Beyond a Single Mapping-Latent Context Drives Time Series Forecastingcited by 1
  5. 2025FAIM: Frequency-Aware Interactive Mamba for Time Series Classificationcited by 0
  6. 2026MLOW: Interpretable Low-Rank Frequency Magnitude Decomposition of Multiple Effects for Time Series Forecastingcited by 0
  7. 2026DySCo: Dynamic Semantic Compression for Effective Long-term Time Series Forecastingcited by 0
  8. 2026Frequency-aware decomposition learning for sensorless wrench estimation in vibration-rich robotic contactcited by 0
  9. 2026Beyond Uniform Tokens: Adaptive Compression for Time Series Language Modelscited by 0
  10. 2026DecoVAE: a Lightweight Interpretable Trend-Seasonal VAE Framework for Efficient Probabilistic Time Series Forecastingcited by 0