Open AccessVehicle Ethernet Flow Estimation Using Intelligent Methods
Author(s) -
Zheng-wei Hu,
Hui Li,
Li Yang,
Xuezhen Li,
Peng Li
Publication year - 2021
Publication title -
journal of physics. conference series
Language(s) - English
Resource type - Journals
SCImago Journal Rank - 0.21
H-Index - 85
eISSN - 1742-6596
pISSN - 1742-6588
DOI - 10.1088/1742-6596/1754/1/012087
Subject(s) - ethernet , computer science , artificial neural network , scheme (mathematics) , ethernet flow control , real time computing , support vector machine , embedded system , computer network , artificial intelligence , mathematical analysis , mathematics
For modern vehicle communications systems, it is vital to utilize advanced networks such as Ethernet to enhance the network speed, improve the service, and avoid the congestion. To this end, it is necessary to develop a hardware platform as well as a flow estimation scheme for system optimization. In this paper, we formulate the vehicle Ethernet platform using Duagon cards under the standard of IEC 61375. Then, we propose three intelligent methods for vehicle flow estimation, i.e., the support vector machine (SVM), the back propagation (BP) neural network, and recurrent neural network (RNN) using improved back-propagation through time (BPTT). Comparative studies are performed to validate the proposed scheme. The results provide technical support for statistical characters analysis and high speed monitoring of modern vehicles.