An Anomaly Node Detection Method for Wireless Sensor Networks Based on Deep Metric Learning with Fusion of Spatial–Temporal Features

Wireless sensor networks (WSNs) use distributed nodes for tasks such as environmental monitoring and surveillance. The existing anomaly detection methods fail to fully capture correlations in multi-node, multi-modal time series data, limiting their effectiveness. Additionally, they struggle with sma...

Full description

Saved in:
Bibliographic Details
Main Authors: Ziheng Wang, Miao Ye, Jin Cheng, Cheng Zhu, Yong Wang
Format: Article
Language:English
Published: MDPI AG 2025-05-01
Series:Sensors
Subjects:
Online Access:https://www.mdpi.com/1424-8220/25/10/3033
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1850126520565104640
author Ziheng Wang
Miao Ye
Jin Cheng
Cheng Zhu
Yong Wang
author_facet Ziheng Wang
Miao Ye
Jin Cheng
Cheng Zhu
Yong Wang
author_sort Ziheng Wang
collection DOAJ
description Wireless sensor networks (WSNs) use distributed nodes for tasks such as environmental monitoring and surveillance. The existing anomaly detection methods fail to fully capture correlations in multi-node, multi-modal time series data, limiting their effectiveness. Additionally, they struggle with small sample scenarios because they do not effectively map features to classes. To address these challenges, this paper presents an anomaly detection approach that integrates deep learning with metric learning. A framework incorporating a graph attention network (GAT) and a Transformer is developed to capture spatial and temporal features. A novel distance measurement module improves similarity learning by considering both intra-class and inter-class relationships. Joint metric-classification training improves model accuracy and generalization. Experiments conducted on public datasets demonstrate that the proposed approach achieves an F1 score of 0.89, outperforming the existing approaches by 7%.
format Article
id doaj-art-09eded9907aa4a37b47a0736763e97b4
institution OA Journals
issn 1424-8220
language English
publishDate 2025-05-01
publisher MDPI AG
record_format Article
series Sensors
spelling doaj-art-09eded9907aa4a37b47a0736763e97b42025-08-20T02:33:55ZengMDPI AGSensors1424-82202025-05-012510303310.3390/s25103033An Anomaly Node Detection Method for Wireless Sensor Networks Based on Deep Metric Learning with Fusion of Spatial–Temporal FeaturesZiheng Wang0Miao Ye1Jin Cheng2Cheng Zhu3Yong Wang4School of Information and Communication, Guilin University of Electronic Technology, Guilin 541000, ChinaSchool of Information and Communication, Guilin University of Electronic Technology, Guilin 541000, ChinaSchool of Information and Communication, Guilin University of Electronic Technology, Guilin 541000, ChinaSchool of Information and Communication, Guilin University of Electronic Technology, Guilin 541000, ChinaSchool of Computer and Information Security, Guilin University of Electronic Technology, Guilin 541000, ChinaWireless sensor networks (WSNs) use distributed nodes for tasks such as environmental monitoring and surveillance. The existing anomaly detection methods fail to fully capture correlations in multi-node, multi-modal time series data, limiting their effectiveness. Additionally, they struggle with small sample scenarios because they do not effectively map features to classes. To address these challenges, this paper presents an anomaly detection approach that integrates deep learning with metric learning. A framework incorporating a graph attention network (GAT) and a Transformer is developed to capture spatial and temporal features. A novel distance measurement module improves similarity learning by considering both intra-class and inter-class relationships. Joint metric-classification training improves model accuracy and generalization. Experiments conducted on public datasets demonstrate that the proposed approach achieves an F1 score of 0.89, outperforming the existing approaches by 7%.https://www.mdpi.com/1424-8220/25/10/3033wireless sensor networksanomaly detectiongraph neural networkmetric learning
spellingShingle Ziheng Wang
Miao Ye
Jin Cheng
Cheng Zhu
Yong Wang
An Anomaly Node Detection Method for Wireless Sensor Networks Based on Deep Metric Learning with Fusion of Spatial–Temporal Features
Sensors
wireless sensor networks
anomaly detection
graph neural network
metric learning
title An Anomaly Node Detection Method for Wireless Sensor Networks Based on Deep Metric Learning with Fusion of Spatial–Temporal Features
title_full An Anomaly Node Detection Method for Wireless Sensor Networks Based on Deep Metric Learning with Fusion of Spatial–Temporal Features
title_fullStr An Anomaly Node Detection Method for Wireless Sensor Networks Based on Deep Metric Learning with Fusion of Spatial–Temporal Features
title_full_unstemmed An Anomaly Node Detection Method for Wireless Sensor Networks Based on Deep Metric Learning with Fusion of Spatial–Temporal Features
title_short An Anomaly Node Detection Method for Wireless Sensor Networks Based on Deep Metric Learning with Fusion of Spatial–Temporal Features
title_sort anomaly node detection method for wireless sensor networks based on deep metric learning with fusion of spatial temporal features
topic wireless sensor networks
anomaly detection
graph neural network
metric learning
url https://www.mdpi.com/1424-8220/25/10/3033
work_keys_str_mv AT zihengwang ananomalynodedetectionmethodforwirelesssensornetworksbasedondeepmetriclearningwithfusionofspatialtemporalfeatures
AT miaoye ananomalynodedetectionmethodforwirelesssensornetworksbasedondeepmetriclearningwithfusionofspatialtemporalfeatures
AT jincheng ananomalynodedetectionmethodforwirelesssensornetworksbasedondeepmetriclearningwithfusionofspatialtemporalfeatures
AT chengzhu ananomalynodedetectionmethodforwirelesssensornetworksbasedondeepmetriclearningwithfusionofspatialtemporalfeatures
AT yongwang ananomalynodedetectionmethodforwirelesssensornetworksbasedondeepmetriclearningwithfusionofspatialtemporalfeatures
AT zihengwang anomalynodedetectionmethodforwirelesssensornetworksbasedondeepmetriclearningwithfusionofspatialtemporalfeatures
AT miaoye anomalynodedetectionmethodforwirelesssensornetworksbasedondeepmetriclearningwithfusionofspatialtemporalfeatures
AT jincheng anomalynodedetectionmethodforwirelesssensornetworksbasedondeepmetriclearningwithfusionofspatialtemporalfeatures
AT chengzhu anomalynodedetectionmethodforwirelesssensornetworksbasedondeepmetriclearningwithfusionofspatialtemporalfeatures
AT yongwang anomalynodedetectionmethodforwirelesssensornetworksbasedondeepmetriclearningwithfusionofspatialtemporalfeatures