Search

As an important part of the new automobile four modernizations, wireless power transfer (WPT) system solves a series of problems of the conductive charging way, and its application is better for the accelerating development of electric vehicles (EVs). At the same time, the electromagnetic safety problem of WPT system also gradually becomes one of the focus of public attention. However, due to the imperfection of relevant standards and regulations, the wireless charging system of electric vehicles is currently in the state of prototype and demonstration operation. The standardized wireless charging system will play a positive role in promoting the large-scale electrification of the wireless charging system of electric vehicles. This paper focuses on the research of EMC of the wireless charging system of electric vehicles. Firstly, based on the structure of the wireless charging system, the possible EMC problems are introduced. Furthermore, the existing EMC standards of wireless charging at home and abroad are compared and analyzed, and the test layout is introduced. Relevant content will provide guidance for enterprises in the development and testing of wireless charging systems.

In this paper, various types of Ethernet harness are taken as the research object, and the transmission characteristic parameters of the harness are measured by network analyzer in real time. The corresponding relationship between the standard high current injection method (BCI), BCI injection harness method in complex electromagnetic environment and the characteristic parameters of the harness are studied. The results show that the change of transmission characteristic parameters of different types of wire harness is closely related to its own length. At the same time, the feasibility of standard electromagnetic interference signal, external electromagnetic environment signal and transmission characteristic parameter change of harness itself is verified.

Aiming at the problem that the precision of gaze information data extraction is not high in the process of pilot's; flight, this paper puts forward the research on the fusion of the data of the landing point of sight and the target information, and feeds back the information through human-computer interaction to obtain the accurate target information. First, the non wearable eye tracker is used to track the line of sight, estimate the fixation point, obtain the fixation point coordinates, and match with the position coordinates of the instrument panel to get the dial where the fixation point is. Then, data fusion is carried out between the instrument panel where the fixation point is located and the flight parameters to obtain the information displayed on the instrument panel. Finally, the fixation information of the instrument panel is used as the input data to support the identification of the pilot's intention. Secondly, data fusion is carried out between the pilot's attention information and reaction time information, and a large amount of data is used as the input data. Machine learning is adopted Methods after information fusion, the decision- making of pilots was predicted.

The iris recognition of the human eye is mainly aimed at iris matching, and there are congenital differences in different iris. In this paper, the iris image is identified by designing efficient convolution neural network(CNN) model. Firstly, with the preprocessing of iris image, the iris region is extracted and divided into eight rectangular sub-regions. And then, those sub-regions are put into the designed model, which can extract the abstract features of the iris automatically. Finally, The iris is identified by the model that learned by a large number of training data. Experiments show that the correctness of the proposed method can reach about 99.0%, which can meet the requirement of iris recognition effectively.

Nowadays, there are more and more hybrid apps appearing in the app market which contain native code and Web pages. In order to enhance the ability of JavaScript in the WebView, these apps expose methods that can be invoked by JavaScript. However, when we study the communication from JavaScript to native code, we find a security issue that if the exposed methods finally invoke sensitive methods, such as SEND SMS, getLastKnownLocation, and these exposed methods are called via unsafe connections, malicious code can be injected to perform sensitive operations without the user's consent. To automatically detect this vulnerability, we provide a hybrid system that contains both static and dynamic analysis modules. The static analysis discerns potential vulnerable apps and gathers information to guide the dynamic analysis while the dynamic analysis executes the app to verify whether the app is vulnerable or not. We use this system to test 400 most popular apps in the Google Play market and find that 43 apps are vulnerable.

Development of the Smart City has produced much data with attributions of timestamp and location, but in some applications like investigation of the large bomb explosion in New York, the government takes precedence to investigate the relation data from New York city rather than the whole Country, which prompts us to do some research works in computing partial graph more fast. So we propose SpatialGraphx, a graph parallel computing framework supporting direct and fast partial graph construction and partial graph computation. Leveraging the spatial and temporal attributions of data, SpatialGraphx presents two extensions on the partial graph construction by building a spatio-temporal tree index and on the computation by a new location-based partition strategy. Using mobile network's data with hundred million edges, we demonstrate SpatialGraphx can support direct and fast partial graph construction and enables efficient partial graph analysis for spatial and temporal data. And compared to original Graphx, the improvement of SpatialGraphx is 3x to several orders of magnitude for large enough dataset.

In order to evaluate the goodness of frequency-hopping sequences, the maximum aperiodic Hamming correlation is used as an important measure. In this paper, based on generalized Gordon-Mills-Welch sequences (GGMW sequences) and generalized m-sequences (GM sequences), two classes of frequency-hopping sequence sets arc presented, which are optimal for maximum aperiodic Hamming correlation according to the aperiodic Hamming correlation bound.

By analyzing the distinguishing features of the current popular weapons fault diagnosis expert system, as well as using component-based software design and complex knowledge representation methods, this paper proposes a solution of overall design and a description of user interfaces about the Weapons Fault Diagnosis Expert System Platform. It studies and discusses the complex knowledge representation about the reusable expert system. The implementation essentials in a variety of reasoning mechanisms are also discussed. It is proved that, a special weapons fault diagnosis expert system can be generated using this expert system platform and the special knowledge of the weapons fault.