Abstract
At present, the train-ground communication system based on the wireless communication protocol is a very important component of communication-based train control (CBTC) systems in intelligent transportation. Its information security is worthy of attention. In order to guarantee the security of the train-ground communication system, this paper proposes an improved AdaBoost multi-classification intrusion detection method based on the n-gram model. First, the n-gram model is used to model the state transitions of the IEEE 802.11 protocol. Then, a typical normal behavior set and typical abnormal behavior sets are obtained by learning and they can portray typical behaviors of their respective classes. Furthermore, a similarity measure algorithm is proposed to construct AdaBoost weak classifiers, which improves the classification effect of AdaBoost algorithm. At last, an AdaBoost multi-classification algorithm is presented to detect and identify the attacks. Experiments prove that the algorithm can effectively detect and distinguish attack types in the train-ground communication system.
Highlights
With the development of city scale and economic level, there are increasingly higher demands for punctuality, energyefficiency, comfort and security of public transportation
The train-ground communication data set was collected from the Beijing Subway Line 7 simulation platform, including normal and attack data
In a detection period, S denotes the number of elements and M denotes the number of the data types
Summary
State Key Laboratory of Rail Traffic Control and Safety, Beijing Jiaotong University, Beijing 100044, China.
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have