Skip to main navigation Skip to search Skip to main content

FusAD: Time-Frequency Fusion with Adaptive Denoising for General Time Series Analysis

  • Northwestern Polytechnical University Xian
  • China Telecommunications
  • University of Science and Technology of China

Research output: Chapter in Book/Report/Conference proceeding › Conference contribution › peer-review

7 Scopus citations

Abstract

Time series analysis plays a vital role in fields such as finance, healthcare, industry, and meteorology, underpinning key tasks including classification, forecasting, and anomaly detection. Although deep learning models have achieved remarkable progress in these areas in recent years, constructing an efficient, multi-task compatible, and generalizable unified framework for time series analysis remains a significant challenge. Existing approaches are often tailored to single tasks or specific data types, making it difficult to simultaneously handle multi-task modeling and effectively integrate information across diverse time series types. Moreover, real-world data are often affected by noise, complex frequency components, and multi-scale dynamic patterns, which further complicate robust feature extraction and analysis. To ameliorate these challenges, we propose FusAD, a unified analysis framework designed for diverse time series tasks. FusAD features an adaptive time-frequency fusion mechanism, integrating both Fourier and Wavelet transforms to efficiently capture global-local and multi-scale dynamic features. With an adaptive denoising mechanism, FusAD automatically senses and filters various types of noise, highlighting crucial sequence variations and enabling robust feature extraction in complex environments. In addition, the framework integrates a general information fusion and decoding structure, combined with masked pre-training, to promote efficient learning and transfer of multigranularity representations. Extensive experiments demonstrate that FusAD consistently outperforms state-of-the-art models on mainstream time series benchmarks for classification, forecasting, and anomaly detection tasks, while maintaining high efficiency and scalability.

Original languageEnglish
Title of host publicationProceedings - 2026 IEEE 42nd International Conference on Data Engineering, ICDE 2026
EditorsKhuzaima Daudjee, Qiong Luo, Alkis Simitsis
PublisherInstitute of Electrical and Electronics Engineers Inc.
Pages1334-1348
Number of pages15
ISBN (Electronic)9798331583651
DOIs
StatePublished - 2026
Event42nd International Conference on Data Engineering, ICDE 2026 - Montreal, Canada
Duration: 4 May 2026 → …

Publication series

NameProceedings - 2026 IEEE 42nd International Conference on Data Engineering, ICDE 2026

Conference

Conference42nd International Conference on Data Engineering, ICDE 2026
Country/TerritoryCanada
CityMontreal
Period4/05/26 → …

Keywords

  • Adaptive Denoising
  • Time Series Analysis
  • Time-Frequency Fusion

Fingerprint

Dive into the research topics of 'FusAD: Time-Frequency Fusion with Adaptive Denoising for General Time Series Analysis'. Together they form a unique fingerprint.

Cite this