Unsupervised Deep Learning for IoT Time Series

TL;DR

This paper reviews unsupervised deep learning methods for IoT time series analysis, focusing on anomaly detection and clustering, and discusses challenges, datasets, and future directions in this emerging field.

Contribution

It provides a systematic survey of unsupervised deep learning techniques applied to IoT time series, filling a gap in existing literature.

Findings

Highlights the effectiveness of DL in feature extraction for IoT data

Identifies key challenges in unsupervised IoT time series analysis

Outlines future research directions and datasets

Abstract

IoT time series analysis has found numerous applications in a wide variety of areas, ranging from health informatics to network security. Nevertheless, the complex spatial temporal dynamics and high dimensionality of IoT time series make the analysis increasingly challenging. In recent years, the powerful feature extraction and representation learning capabilities of deep learning (DL) have provided an effective means for IoT time series analysis. However, few existing surveys on time series have systematically discussed unsupervised DL-based methods. To fill this void, we investigate unsupervised deep learning for IoT time series, i.e., unsupervised anomaly detection and clustering, under a unified framework. We also discuss the application scenarios, public datasets, existing challenges, and future research directions in this area.