Abstract
The controller area networks (CAN) in the braking control system of metro trains are used to transmit the important control instruction and condition information, whose anomaly will endanger the security of trains running seriously. Due to the harsh work environment, there are various known and previously unknown fault types, current scheduled maintenance cannot detect early anomaly in time, and constructing an accurate and stable anomaly detector is a challenging task. In this paper, an anomaly detection approach is proposed to detect anomaly based on a dynamic ensemble selection system (DESS) without the expert knowledge, which involves two-class and one-class classifiers, and the base classifiers are trained with the network features extracted from the physical-layer information. To conduct the fusion, the support function of “distance-based” classifier is redefined as a class-conditional probability density function, and the source competence of base classifier is estimated by the entropy-based method in validate space and extended to entire decision space using the normalized Gaussian potential function. For different fault types, the competence classifiers are selected and the anomaly detection result is finally achieved by weighted majority voting. The comparative experiments are included in this paper to demonstrate the effectiveness and robustness in anomaly detection, including varying fault types.
Highlights
The braking control system is an important part of metro trains and relative to the running security
The dynamic ensemble selection system (DESS) for anomaly detection method consists of four successive steps: first, since the physical-layer of the controller area networks (CAN) protocol contains most of the network performance and failure information [10]–[13], feature sets extracted from the physical-layer information are divided into separate training, validation, and testing sets in different fault types
THE METHODOLOGY: DYNAMIC ENSEMBLE SELECTION SYSTEM FOR ANOMALY DETECTION Based on dynamic ensemble learning, this study proposes an intelligent anomaly detection approach that makes use of the feature characteristics from the physical-layer information and detects the known and unknown fault
Summary
The braking control system is an important part of metro trains and relative to the running security. Through the use of a labeled set contains normal and abnormal behavior for training, an anomaly detection system is obtained to classify a test set as either normal or anomaly. This corresponds to supervised anomaly detection which has been reported to work for CAN bus [3]. The DESS for anomaly detection method consists of four successive steps: first, since the physical-layer of the CAN protocol contains most of the network performance and failure information [10]–[13], feature sets extracted from the physical-layer information are divided into separate training, validation, and testing sets in different fault types.
Sign-in/Register to view highlights & summary Sign-in/Register to access full text options 
Free) 
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have
Schedule a call
