Machine Learning-Based Anomaly Detection of Correlated Sensor Data: An Integrated Principal Component Analysis-Autoencoder Approach

This research proposes a lightweight hybrid approach for anomaly detection in correlated IoT sensor data, combining PCA for fast monitoring and Autoencoders for deeper analysis. Validated on real and simulated data, the method offers high accuracy, faster response, and fewer false positives-ideal fo...

Full description

Saved in:
Bibliographic Details
Published inProceedings of the ... IEEE Workshop on Local and Metropolitan Area Networks pp. 1 - 2
Main Authors Baranwal, Tanish, Das, Arnab, Varada, Srihari, Das, Santanu, Haider, Mohammad R.
Format Conference Proceeding
LanguageEnglish
Published IEEE 07.07.2025
Subjects
Accuracy
Anomaly detection
Autoencoder
Autoencoders
correlated sensor data
false positives
IoT
machine learning
Metropolitan area networks
Mission critical systems
Monitoring
PCA
Principal component analysis
Real-time systems
Robustness
Time factors
Online AccessGet full text
ISSN1944-0375
DOI10.1109/LANMAN66415.2025.11154513

Cover

More Information
Summary:This research proposes a lightweight hybrid approach for anomaly detection in correlated IoT sensor data, combining PCA for fast monitoring and Autoencoders for deeper analysis. Validated on real and simulated data, the method offers high accuracy, faster response, and fewer false positives-ideal for resource-constrained environments.
ISSN:1944-0375
DOI:10.1109/LANMAN66415.2025.11154513