Anomaly Detection for Non-Stationary and Non-Periodic Univariate Time Series

Yu Lin Li, Jehn Ruey Jiang

Research output: Chapter in Book/Report/Conference proceedingConference contributionpeer-review

10 Scopus citations

Abstract

This study proposes an anomaly detection method called wavelet autoencoder anomaly detection (WAAD) for non-stationary and non-periodic univariate time series. The method first applies discrete wavelet transform to time series of a sliding time window to obtain wavelet transform coefficients. It then uses an autoencoder to encode and decode (reconstruct) these coefficients. WAAD calculates the reconstruction error for every time window. An anomaly is assumed to occur for specific conditions of the errors. By five NAB datasets, the performance of WAAD is evaluated and compared with other methods to show its superiority.

Original languageEnglish
Title of host publication2nd IEEE Eurasia Conference on IOT, Communication and Engineering 2020, ECICE 2020
EditorsTeen-Hang Meen
PublisherInstitute of Electrical and Electronics Engineers Inc.
Pages177-179
Number of pages3
ISBN (Electronic)9781728180601
DOIs
StatePublished - 23 Oct 2020
Event2nd IEEE Eurasia Conference on IOT, Communication and Engineering, ECICE 2020 - Yunlin, Taiwan
Duration: 23 Oct 202025 Oct 2020

Publication series

Name2nd IEEE Eurasia Conference on IOT, Communication and Engineering 2020, ECICE 2020

Conference

Conference2nd IEEE Eurasia Conference on IOT, Communication and Engineering, ECICE 2020
Country/TerritoryTaiwan
CityYunlin
Period23/10/2025/10/20

Keywords

  • anomaly detection
  • autoencoder
  • time series
  • wavelet transform

Fingerprint

Dive into the research topics of 'Anomaly Detection for Non-Stationary and Non-Periodic Univariate Time Series'. Together they form a unique fingerprint.

Cite this