Applied Mechanics and Materials Vols. 347-350

Abstract: Due to the implementation of information technology and the various e-government new businesses, e-government system has generated a lot of valuable information resources as well as a large amount of redundant information. Information sharing can save the cost of information collection, and the collected information can be used by many a government agency. Based on the sharing of information resources, an e-government system framework is design in the paper at three levels, namely, network infrastructure, information resources, and data exchange platform.

Abstract: Segmentation is generally an important prior process to automatic analysis of heart sounds. The proposed algorithm achieved this in separated cycles. First, the cyclostationary envelopes of heart sound signals are calculated. Second, a threshold is chosen to separate S1 and S2. The segmentation is automatic, and robust to noise. It can identify S1 and S2 correctly even under the circumstance of noise and interference. No reference signal is needed for this segmentation. It was tested by the heart sounds of 20 subjects including 15 normal and 5 abnormal in various clinical cases (715 cycles in total). The statistics showed over 96 percent correctness for segmentation.

Abstract: Database technology is of great variety, structure diversity, solving data integration and exchanges sharing of the heterogeneous database system have became the important problem in all fields. Because of XML is the very important data exchange technology to realize data exchange and transmission ability, this paper mainly explores the data exchange problem how to use XML in heterogeneous database system, , and .Net platform have good support to XML and heterogeneous database system, so the combination of XML and.net, can realize data exchange of heterogeneous database system.

Abstract: Classical percolation processes on random networks always have only one giant cluster after the transition. Recently, percolation with multiple giant clusters has been observed and attracted much attention. In this paper, we show three simple but different models which would lead to multiple giant clusters. The evolution of these models is discussed respectively and they are supported by numerical simulations. The study of these new models may present a deep understanding of the emergence of multiple giant clusters in random networks.

Abstract: Multicast routing is a critical problem in the P2P network. The main task for multicast routing is to establish a multicast tree with good performance, to satisfy the various qualities of service requirements, such as improving the streaming media service quality, and the utilization rate of cyber source. Many efficient overlay multicast routing algorithms are proposed to face the new challenge in current P2P network. This paper attempts to survey the current problems, the solutions for overlay multicast routing algorithms, and points out the further research directions.

Abstract: Based on the integration of space and time, TGIS system is the most important cornerstone for emergency rescue, while the spatial-temporal data model, in turn, is the vital foundation of TGIS. In recent years, various kinds of spatial-temporal data model have been developed. Consequently, this paper is trying to launch a research on the classification of these models and try to propose a discussion of models, including simple models, Space-time composites, models based on object and feature, and models based on object and feature. Meantime, the adaptability of the four types of models is also discussed in the paper, which shall provide reference for emergency space-time sequence and spatial-temporal data mining.

Abstract: In this paper, we present a general and extensible context-aware computing ontology (CACOnt) for modeling context and providing inference mechanisms. CACOnt provides not only the generic context ontologies for capturing basic concepts about context, but also the extensibility for adding domain-specific ontologies in a hierarchical manner. CACOnt facilitates the context reasoning capabilities by providing semantic logics which is possible to combine with rule-based systems. However, the set of rules cannot entirely cover the domain of contexts, we present a semantic similarity-based rule matching algorithm as the solution to this problem.

Abstract: Cloud computing is becoming a popular and important platform for web service applications. With the advent of more and more service resources in cloud, it is a crucial challenge to build an effective diagnostic mechanism. To improve the diagnosis capability for composite web service, we propose an automatic diagnostic framework and a hybrid diagnosis method for diagnosing the faulty activities. Our diagnostic framework can improve the scalability of diagnostic system and add the diagnostic capability of diagnosis service by decoupling the diagnostic service components. Our diagnosis method identifies the differences between successful and failed executions for finding the incorrect activities of the composite service and provides faulty causes by considering the structured activities in the process of diagnosis. Experimental results show that our method is effective to the composite web service in the fault diagnosis.

Abstract: In order to meet the requirements of multiple data links cooperative combat in network centric battle field, we analyzed the problems and challenges that US Army tactical data link system encountered when meeting the needs of network-centric warfare, and elaborated the turning point that the ideas of cloud computing service mode construct the networking data link system in network-centric battle field. On the basis of this, we put forward a new battlefield-oriented, demand-based, real-time serviced networking data link system---the joint data link system based on cloud computing, and proved that the use of cloud technology in Joined Data Links can improve the system service level markedly.

Abstract: Missing data universally exists in various research fields and it results in bad computational performance and effcet. In order to improve the accuracy of filling in the missing data, a filling missing data algorithm of the nearest neighbor based on the cluster analysis is proposed by this paper. After clustering data analysis,the algorithm assigns weights according to the categories and improves calculation formula and filling value calculation based on the MGNN (Mahalanobis-Gray and Nearest Neighbor algorithm) algorithm.The experimental results show that the filling accuracy of the method is higher than traditional KNN algorithm and MGNN algorithm.