Anomaly Detection for Controller Area Network in Braking Control System With Dynamic Ensemble Selection

Yang, Yueyi; Wang, Lide; Li, Zhaozhao; Shen, Peng; Guan, Xingwen; Xia, Wenchao

doi:10.1109/access.2019.2929162

Cited by 16 publications

(3 citation statements)

References 42 publications

(42 reference statements)

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Anomaly detection has been addressed using different approaches namely multivariate statistical methods (e.g., Hoteling, Mahalanobis distance) [29], [30], variational autoencoders (VAE) [31], [32], Bayesian neural networks (BNN) [33], [34], ensemble learning [17], [35], among others. An extensive survey of the different methods for anomaly detection is found in [36].…”

Section: Anomaly Detectionmentioning

confidence: 99%

Anomaly Detection Using Ensemble Classification and Evidence Theory

et al. 2023

View full text Add to dashboard Cite

Multi-class ensemble classification remains a popular focus of investigation within the research community. The popularization of cloud services has sped up their adoption due to the ease of deploying large-scale machine-learning models. It has also drawn the attention of the industrial sector because of its ability to identify common problems in production. However, there are challenges to conform an ensemble classifier, namely a proper selection and effective training of the pool of classifiers, the definition of a proper architecture for multi-class classification, and uncertainty quantification of the ensemble classifier. The robustness and effectiveness of the ensemble classifier lie in the selection of the pool of classifiers, as well as in the learning process. Hence, the selection and the training procedure of the pool of classifiers play a crucial role. An (ensemble) classifier learns to detect the classes that were used during the supervised training. However, when injecting data with unknown conditions, the trained classifier will intend to predict the classes learned during the training. To this end, the uncertainty of the individual and ensemble classifier could be used to assess the learning capability. We present a novel approach for anomaly detection using ensemble classification and evidence theory. A pool selection strategy is presented to build a solid ensemble classifier. We present an architecture for multi-class ensemble classification and an approach to quantify the uncertainty of the base classifiers and the ensemble classifier. We address the problem of detecting unknown conditions (while feeding out-of-distribution data), presenting a novel approach that monitors the uncertainty of the ensemble classifier using evidence theory. Finally, we use the benchmark Tennessee Eastman to perform experiments to test the ensemble classifier's prediction and anomaly detection capabilities.

show abstract

Section: Anomaly Detectionmentioning

confidence: 99%

Anomaly Detection Using Ensemble Classification and Evidence Theory

et al. 2023

View full text Add to dashboard Cite

show abstract

“…In addition to voltage-based IDS, the timing of signal is another characteristics which can be exploited to construct fingerprint for ECUs. The work done by Yang et al [34] employs more comprehensive physical characteristics of signal including both voltage and time. It provides a system ensembled with two kinds of classifier and a total of six classification algorithms to construct stable detector for attacks.…”

Section: B Related Workmentioning

confidence: 99%

Clock-Based Sender Identification and Attack Detection for Automotive CAN Network

Zhou

Xie

et al. 2021

IEEE Access

View full text Add to dashboard Cite

Building the security mechanism for Controller Area Network (CAN) to defend against attack has drawn substantial attention recently. Fingerprinting ECUs to provide the ability of authentication based on the physical characteristics can protect the CAN network effectively. The clock skew which is unique and stable can be exploited to pinpoint the attacker and detect intrusion. However, a common downside of existing clock-skew-based approaches is that the estimation process can be affected by the message scheduling or arbitration. In our work, a novel intrusion detection system (IDS) that exploits the inherent difference in the clock of devices for automotive CAN network is designed. The estimation process of clock skew in our approach relies only on the time measurement of a single CAN frame. Thus, the disturbance from the data-link layer can be avoided. Since the performance of our IDS depends heavily on the accuracy of estimated clock skew, our approach is evaluated on CAN networks with different settings to simulate cases in which the sampling rate is sufficient or not. The feasibility as well as the limitation of our approach are presented in our work. The evaluation shows that our IDS can identify the sender and detect attacks with an average identification rate of more than 99.7% when the sampling rate is sufficient. Besides, the performance degradation as low sampling accuracy is shown and feasible measures for improvement are also discussed.

show abstract

“…Yang et al . have also extracted waveform features from the train braking network . They have proposed a dynamic ensemble selection method to detect the known network faults and unknown network faults.…”

Section: Introductionmentioning

confidence: 99%

Fault diagnosis of the train communication network based on weighted support vector machine

Wang

Yang

2020

IEEJ Transactions Elec Engng

Self Cite

View full text Add to dashboard Cite

Multifunction vehicle bus (MVB) is the most widely used train communication network which transmits controlling and supervising data. The faults of MVB will heavily affect the train's safe and stable operation. Due to the harsh operating environment and distributed structure, the MVB fault diagnosis has always been a difficult issue in the maintenance of the train. Many MVB faults will distort the physical waveforms and cause serious packet loss. Thus, we have extracted waveform features to characterize different MVB faults and turned the fault diagnosis into a pattern recognition problem. Then a classifier based on weighted support vector machine (WSVM) has been trained to diagnose and locate network faults. Considering that samples locating in different positions of the feature space have different influences on the support vector machine (SVM) hyperplane, we have proposed a multi‐hop edge approaching method to assign sample weights in WSVM. To identify the position of the tested sample, the hops from its location to the classification margin have been counted. The less the hop‐count, the closer to the classification margin and the larger the sample weight. Compared with SVM and other WSVM methods, the proposed method has better performance on the artificial synthetic datasets, the MVB dataset, and the benchmark datasets. © 2020 Institute of Electrical Engineers of Japan. Published by John Wiley & Sons, Inc.

show abstract

Anomaly Detection for Controller Area Network in Braking Control System With Dynamic Ensemble Selection

Cited by 16 publications

References 42 publications

Anomaly Detection Using Ensemble Classification and Evidence Theory

Anomaly Detection Using Ensemble Classification and Evidence Theory

Clock-Based Sender Identification and Attack Detection for Automotive CAN Network

Fault diagnosis of the train communication network based on weighted support vector machine

Contact Info

Product

Resources

About