Advanced Materials Research Vol. 940

Abstract: The wireless sensor network localization algorithm in this paper combines hop-count information and distributed learning. The network is classified into many classes based on sensors’ location, and then the class that each sensor falls into is specified. There are a certain number of beacon nodes with position coordinate in network, and they use their own locations as training data in performing above classification. This positioning method merely uses the partial hop-count information between target sensor and reference node in specifying the class of each node. The final simulation experiment will analyze the excellent performance of this method under different system parameters.

Abstract: Trusted Network is a new research direction of the Internet. On the basis of trusted LAN, this study puts forward a trusted label of the trusted terminal, which can solve the DDOS attack efficiently and control the abnormal flow easily in the trusted LAN. In addition, it proposes the label’s verification technology, so as to ensure the execution efficiency of some real-time business in the gateway. Above all, the trusted label and the verification technology make the LAN more controllable and safer.

Abstract: This research proposes network access control model based on the trusted label. The model combines the admission access control of network device and the transmission access control of data flow effective, and uses the trusted computing technology to ensure the credibility of the generated label, and adopts the access control based on the security domain to locate the granularity of access control to every packets, reduces the probability of successful attack to the network based on the information flow, thereby improving the security control of the network.

Abstract: Due to the extensive application of automobile electronic control technologies, the demand of automotive network is growing. Automotive operation data sharing has become a feature of the automotive network. CAN network is most widely applied in the automobiles. LIN network is a supplement for CAN network in practical application. The hybrid network of CAN/LIN can improve the stability of the system, and also reduce network cost without influencing the performance of the network. In this paper, through the study of CAN/LIN hybrid network technologies of automotive body, a system scheme of the automotive body network with CAN/LIN dedicated gateway is proposed.Keyword: CAN Bus; LIN Bus; Bus network technology; CAN/LIN hybrid network

Abstract: Along with the RFID applied on vehicle technology researching, it is very important that the RFID technology and video technology were combined to management the vehicle moving on the road. Under several projects supported by government, we have a good opportunity to experiment and solve the exact matching problem of the RFID and image of moving vehicle. Firstly, the same space-time exact matching method for the RFID and image of moving vehicle under high speed free flow road conditions was proposed. Secondly, the experimental scheme was made by introducing the vehicle sensing coil to bridge the RFID and image information. Lastly, we obtained a large number of experimental data and made a lot of comparison analysis. The experimental results shown that the method is scientific and practical.

Abstract: A novel modeling method based on multi-dimensional Taylor network is proposed. The structure and the principle of the multi-dimensional Taylor network are introduced. Based on this, the method is applied in the nonlinear time series prediction based on multi-dimensional Taylor network. It provides a new method to predict the time series, which can describe the dynamic characteristics without prior knowledge and can realize the prediction of the nonlinear time series just with input-output data. An example of predicting the stress data of a large span bridge tower induced by strong typhoon is taken at last in this paper. Results indicate the validity and the better prediction accuracy of this method in nonlinear time series prediction.

Abstract: As a mathematical model for cyber-physical system, hybrid systems are dynamical systems that are governed by interacting discrete and continuous dynamics. In this paper, we present a symbolic-numeric hybrid method to generate inequality inductive invariants for safety verification of polynomial hybrid systems, and based on this method, we develop an automated verification tool HSProver.

Abstract: DINA model appreciated by many researchers considers the relationships among attributes as mutually independent, conjunction and non-compensation relationship, but in real applications, there may not be able to meet such a relationship. Study the impacts on parameter estimates for the DINA model because of attribute relationships. Simulation results show that when there is a hierarchical relationship between attributes, there will be a great impact on the parameter estimation accuracy of DINA model; and the parameter estimation accuracy mainly affected by attribute hierarchy and the number of subjects. When existing hierarchy relationships among attributes, using DINA model as cognitive diagnosis model would affect the validity of diagnostic tests.

Abstract: This paper researches on a method for on-line fault calculation based on expert system theory. The method improves traditional method of width precedence searching, and applies to form the node impedance matrix, which not only settles loop net but also advances calculation efficiency. To convenient for fault calculation, we process and identify the basic parameters and perform equivalent of 220kV external system. Moreover inbuilt fault calculation software has been realized, which can exchange data with SCADA system on-line.

Abstract: When building the cloud service platform for manufacturing, we find that the cloud platform is not so perfect in the knowledge service pattern that is the reason why we propose to build expert system. While, the traditional expert system is passive in achieving the knowledge, and its reasoning ability is not so good, which can’t adapt to the cloud service pattern. So we want to use neural network expert system to build a new expert system.