Unsupervised Deep Learning for IoT Time Series

Liu, Ya; Zhou, Yingjie; Yang, Kai; Wang, Xin

doi:10.1109/jiot.2023.3243391

Cited by 15 publications

(3 citation statements)

References 179 publications

(234 reference statements)

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Anomaly detection in network traffic data is paramount for cybersecurity. Multivariate time series analysis allows for the identification of subtle deviations from normal patterns, potentially revealing cyberattacks or intrusions [85,86,87]. In industrial plants, analyzing sensor data from machinery can help predict equipment failures before they occur [88,89,90,91].…”

Section: Related Workmentioning

confidence: 99%

Securing the IoT Cyber Environment: Enhancing Intrusion Anomaly Detection With Vision Transformers

Sana,

Nazir,

Yang

et al. 2024

IEEE Access

View full text Add to dashboard Cite

Background and Motivation:The ever-expanding Internet of Things (IoT) landscape presents a double-edged sword. While it fosters interconnectedness, the vast amount of data generated by IoT devices creates a larger attack surface for cybercriminals. Intrusions in these environments can have severe consequences. To combat this growing threat, robust intrusion detection systems (IDS) are crucial. The data comprised by this attack is multivariate, highly complex, non-stationary, and nonlinear. To extract the complex patterns from this complex data, we require the most robust, optimized tools. Methods: Machine learning (ML) and deep learning (DL) have emerged as powerful tools for IDSs, offering high accuracy in detecting and preventing security breaches. This research delves into anomaly detection, a technique that identifies deviations from normal system behavior, potentially indicating attacks. Given the complexity of anomaly data, we explore methods to improve detection performance. Proposed Approach: This research investigates the design and evaluation of a novel IDS. We leverage and optimize supervised ML methods like tree-based Support Vector Machines (SVM), ensemble methods, and neural networks (NN) alongside the cutting-edge DL approach of long short-term memory (LSTM) and vision transformers (ViT). We optimized the hyperparameters of these algorithms using a robust Bayesian optimization approach. Results: The implemented ML models achieved impressive training accuracy, with Random Forest and Ensemble Bagged Tree surpassing 99.90% of accuracy, an AUC of 1.00, an F1-score, and a balanced Matthews Correlation Coefficient (MCC) of 99.78%. While the initial deep learning LSTM model yielded an accuracy of 99.97%, the proposed ViT architecture significantly boosted performance with 100% of all metrics, along with a validation accuracy of 78.70% and perfect training accuracy. Conclusion: Our findings demonstrate the potential of the proposed methods for enhanced attack and intrusion detection. This improved detection capability can be instrumental in safeguarding system integrity, identifying fraudulent activity, and optimizing system performance in IoT networks.

show abstract

Section: Related Workmentioning

confidence: 99%

Securing the IoT Cyber Environment: Enhancing Intrusion Anomaly Detection With Vision Transformers

Sana,

Nazir,

Yang

et al. 2024

IEEE Access

View full text Add to dashboard Cite

show abstract

“…Deep learning (DL) has recently made internet of things (IoT)-based multivariate time-series analysis possible because of its potent feature extraction and representation learning capabilities. Nevertheless, some existing time-series analysis studies have included unsupervised DL-based techniques [3]. To close this knowledge gap, we examine unsupervised learning-based risk detection and clustering for IoT time series within a unified framework [4,5].…”

Section: Introductionmentioning

confidence: 99%

NDAMA: A Novel Deep Autoencoder and Multivariate Analysis Approach for IoT-Based Methane Gas Leakage Detection

Dashdondov,

Kim,

2023

IEEE Access

View full text Add to dashboard Cite

Natural gas is widely used for domestic and industrial purposes, and whether it is being leaked into the air cannot be directly known. The current problem is that gas leakage is not only economically harmful but also detrimental to health. Therefore, much research has been done on gas damage and leakage risks, but research on predicting gas leakages is just beginning. In this study, we propose a method based on deep learning to predict gas leakage from environmental data. Our proposed method has successfully improved the performance of machine learning classification algorithms by efficiently preparing training data using a deep autoencoder model. The proposed method was evaluated on an open dataset containing natural gas and environmental information and compared with extreme gradient boost (XGBoost), K-nearest neighbors (KNN), decision tree (DT), random forest (RF), and naive Bayes (NB) algorithms. The proposed method is evaluated using accuracy, F1-score, mean square error (MSE), mean intersection over union (mIoU), and area under the ROC curve (AUC). The presented method in this study outperformed all compared methods. Moreover, the deep autoencoder and ordinal encoder-based XGBoost (DA-MA-XGBoost) showed the best performance by giving 99.51% accuracy, an F1-score of 99.53%, an MSE of 0.003, mIoU of 99.40 and an AUC of 99.62%.INDEX TERMS Deep autoencoder, multivariate analysis, methane gas, risk detection, XGBoost.

show abstract

“…Supervised, semi-supervised, and unsupervised paradigms are the most utilized for AD [26]. In this paper, we specifically focus on unsupervised methods as they are able to automatically discern anomalies without any external supervision or labeled data [26], [34].…”

Section: Introductionmentioning

confidence: 99%

A Tiny Transformer-Based Anomaly Detection Framework for IoT Solutions

Barbieri,

Brambilla,

Stefanutti

et al. 2023

IEEE Open J. Signal Process.

View full text Add to dashboard Cite

This paper addresses the problem of multitarget tracking using a network of sensing agents with unknown positions. Agents have to both localize themselves in the sensor network and, at the same time, perform multitarget tracking in the presence of clutter and miss detection. These two problems are jointly resolved using a holistic and centralized approach where graph theory is used to describe the statistical relationships among agent states, target states, and observations. A scalable message passing scheme, based on the sum-product algorithm, enables to efficiently approximate the marginal posterior distributions of both agent and target states. The proposed method is general enough to accommodate a full multistatic network configuration, with multiple transmitters and receivers. Numerical simulations show superior performance of the proposed joint approach with respect to the case in which cooperative self-localization and multitarget tracking are performed separately, as the former manages to extract valuable information from targets. Lastly, data acquired in 2018 by the NATO Science and Technology Organization (STO) Centre for Maritime Research and Experimentation (CMRE) through a network of autonomous underwater vehicles demonstrates the effectiveness of the approach in a practical application.

show abstract

Unsupervised Deep Learning for IoT Time Series

Cited by 15 publications

References 179 publications

Securing the IoT Cyber Environment: Enhancing Intrusion Anomaly Detection With Vision Transformers

Securing the IoT Cyber Environment: Enhancing Intrusion Anomaly Detection With Vision Transformers

NDAMA: A Novel Deep Autoencoder and Multivariate Analysis Approach for IoT-Based Methane Gas Leakage Detection

A Tiny Transformer-Based Anomaly Detection Framework for IoT Solutions

Contact Info

Product

Resources

About