Enhancing IoT anomaly detection performance for federated learning
Abstract
Federated Learning (FL) with mobile computing and the Internet of Things (IoT) is an effective cooperative learning approach. However, several technical challenges still need to be addressed. For instance, dividing the training process among several devices may impact the performance of Machine Learning (ML) algorithms, often significantly degrading prediction accuracy compared to centralized learning. One of the primary reasons for such performance degradation is that each device can access only a small fraction of data (that it generates), which limits the efficacy of the local ML model constructed on that device. The performance degradation could be exacerbated when the participating devices produce different classes of events, which is known as the class balance problem. Moreover, if the participating devices are of different types, each device may never observe the same types of events, which leads to the device heterogeneity problem. In this study, we investigate how data augmentation can be applied to address these challenges and improving detection performance in an anomaly detection task using IoT datasets. Our extensive experimental results with three publicly accessible IoT datasets show the performance improvement of up to 22.9% with the approach of data augmentation, compared to the baseline (without relying on data augmentation).more »
- Publication Date:
- Research Org.:
- Lawrence Berkeley National Lab. (LBNL), Berkeley, CA (United States)
- Sponsoring Org.:
- USDOE Office of Science (SC), Advanced Scientific Computing Research (ASCR)
- OSTI Identifier:
- 1859795
- Alternate Identifier(s):
- OSTI ID: 1894087
- Grant/Contract Number:
- AC02-05CH11231
- Resource Type:
- Published Article
- Journal Name:
- Digital Communications and Networks
- Additional Journal Information:
- Journal Name: Digital Communications and Networks Journal Volume: 8 Journal Issue: 3; Journal ID: ISSN 2352-8648
- Publisher:
- Elsevier
- Country of Publication:
- Netherlands
- Language:
- English
- Subject:
- 97 MATHEMATICS AND COMPUTING; data augmentation; federated learning; Internet of things; anomaly detection; machine learning
Citation Formats
Weinger, Brett, Kim, Jinoh, Sim, Alex, Nakashima, Makiya, Moustafa, Nour, and Wu, K. John. Enhancing IoT anomaly detection performance for federated learning. Netherlands: N. p., 2022.
Web. doi:10.1016/j.dcan.2022.02.007.
Weinger, Brett, Kim, Jinoh, Sim, Alex, Nakashima, Makiya, Moustafa, Nour, & Wu, K. John. Enhancing IoT anomaly detection performance for federated learning. Netherlands. https://doi.org/10.1016/j.dcan.2022.02.007
Weinger, Brett, Kim, Jinoh, Sim, Alex, Nakashima, Makiya, Moustafa, Nour, and Wu, K. John. Wed .
"Enhancing IoT anomaly detection performance for federated learning". Netherlands. https://doi.org/10.1016/j.dcan.2022.02.007.
@article{osti_1859795,
title = {Enhancing IoT anomaly detection performance for federated learning},
author = {Weinger, Brett and Kim, Jinoh and Sim, Alex and Nakashima, Makiya and Moustafa, Nour and Wu, K. John},
abstractNote = {Federated Learning (FL) with mobile computing and the Internet of Things (IoT) is an effective cooperative learning approach. However, several technical challenges still need to be addressed. For instance, dividing the training process among several devices may impact the performance of Machine Learning (ML) algorithms, often significantly degrading prediction accuracy compared to centralized learning. One of the primary reasons for such performance degradation is that each device can access only a small fraction of data (that it generates), which limits the efficacy of the local ML model constructed on that device. The performance degradation could be exacerbated when the participating devices produce different classes of events, which is known as the class balance problem. Moreover, if the participating devices are of different types, each device may never observe the same types of events, which leads to the device heterogeneity problem. In this study, we investigate how data augmentation can be applied to address these challenges and improving detection performance in an anomaly detection task using IoT datasets. Our extensive experimental results with three publicly accessible IoT datasets show the performance improvement of up to 22.9% with the approach of data augmentation, compared to the baseline (without relying on data augmentation). In particular, stratified random sampling and uniform random sampling show the best improvement in detection performance with only a modest increase in computation time, whereas the data augmentation scheme using Generative Adversarial Networks is the most time-consuming with limited performance benefits.},
doi = {10.1016/j.dcan.2022.02.007},
journal = {Digital Communications and Networks},
number = 3,
volume = 8,
place = {Netherlands},
year = {2022},
month = {6}
}
Free Publicly Available Full Text
Publisher's Version of Record
https://doi.org/10.1016/j.dcan.2022.02.007
https://doi.org/10.1016/j.dcan.2022.02.007
Other availability
Search WorldCat to find libraries that may hold this journal
Save to My Library
You must Sign In or Create an Account in order to save documents to your library.
Works referenced in this record:
IoT Privacy and Security Challenges for Smart Home Environments
journal, July 2016
- Lin, Huichen; Bergmann, Neil
- Information, Vol. 7, Issue 3
Chained Anomaly Detection Models for Federated Learning: An Intrusion Detection Case Study
journal, December 2018
- Preuveneers, Davy; Rimmer, Vera; Tsingenopoulos, Ilias
- Applied Sciences, Vol. 8, Issue 12
Capabilities and limitations of wireless CO2, temperature and relative humidity sensors
journal, May 2019
- Mylonas, Angelos; Kazanci, Ongun Berk; Andersen, Rune K.
- Building and Environment, Vol. 154
Secure Medical Data Transmission Model for IoT-Based Healthcare Systems
journal, January 2018
- Elhoseny, Mohamed; Ramirez-Gonzalez, Gustavo; Abu-Elnasr, Osama M.
- IEEE Access, Vol. 6
A study of the stratified random sampling
journal, December 1954
- Aoyama, Hirojiro
- Annals of the Institute of Statistical Mathematics, Vol. 6, Issue 1
N-BaIoT—Network-Based Detection of IoT Botnet Attacks Using Deep Autoencoders
journal, July 2018
- Meidan, Yair; Bohadana, Michael; Mathov, Yael
- IEEE Pervasive Computing, Vol. 17, Issue 3
PrivacyProtector: Privacy-Protected Patient Data Collection in IoT-Based Healthcare Systems
journal, February 2018
- Luo, Entao; Bhuiyan, Md Zakirul Alam; Wang, Guojun
- IEEE Communications Magazine, Vol. 56, Issue 2
Generative adversarial networks
journal, October 2020
- Goodfellow, Ian; Pouget-Abadie, Jean; Mirza, Mehdi
- Communications of the ACM, Vol. 63, Issue 11
User Perceptions of Smart Home IoT Privacy
journal, November 2018
- Zheng, Serena; Apthorpe, Noah; Chetty, Marshini
- Proceedings of the ACM on Human-Computer Interaction, Vol. 2, Issue CSCW
Data augmentation using generative adversarial networks (CycleGAN) to improve generalizability in CT segmentation tasks
journal, November 2019
- Sandfort, Veit; Yan, Ke; Pickhardt, Perry J.
- Scientific Reports, Vol. 9, Issue 1
Anomaly detection: A survey
journal, July 2009
- Chandola, Varun; Banerjee, Arindam; Kumar, Vipin
- ACM Computing Surveys, Vol. 41, Issue 3, p. 1-58
Attack and anomaly detection in IoT sensors in IoT sites using machine learning approaches
journal, September 2019
- Hasan, Mahmudul; Islam, Md. Milon; Zarif, Md Ishrak Islam
- Internet of Things, Vol. 7
Deep Learning for Anomaly Detection: A Review
journal, March 2022
- Pang, Guansong; Shen, Chunhua; Cao, Longbing
- ACM Computing Surveys, Vol. 54, Issue 2
Efficient IoT-based sensor BIG Data collection–processing and analysis in smart buildings
journal, May 2018
- Plageras, Andreas P.; Psannis, Kostas E.; Stergiou, Christos
- Future Generation Computer Systems, Vol. 82
Fog-Empowered Anomaly Detection in IoT Using Hyperellipsoidal Clustering
journal, October 2017
- Lyu, Lingjuan; Jin, Jiong; Rajasegarar, Sutharshan
- IEEE Internet of Things Journal, Vol. 4, Issue 5
Federated Learning: Challenges, Methods, and Future Directions
journal, May 2020
- Li, Tian; Sahu, Anit Kumar; Talwalkar, Ameet
- IEEE Signal Processing Magazine, Vol. 37, Issue 3