Research And Implementation Of Core Technologies Of Moving Objects Database In Networks

Moving objects databases mainly deal with the objects moving continuously with time and the core research work is the storeage structure and toplogical operators. The databases of moving objects in networks which is on the basis of spatial databases and spatio-temporal databases mainly study the objects moving in constraint environment. The aim of researching moving objects in networks is to provide a reliable platform of artificial transport system for transportation analyzing and decision-making. This paper focuses on several core issues and technologies of moving objects in networks and the storing structure of moving objects databases, the main contributions are as follows:First of all, it improves the current network model including data types and operators so that it can include multi-roads, the control of signal lights which is the basis of the optimal trip query, and the types and attitudes of moving objects. In addition, it implements all the data types and 90 operators of moving objects in networks in an extensible database system-Secondo.Secondly, for the low efficiency and drawback of current storing structure and operators in moving objects databases, it presents a two-level storing structure TLSS. It can store and manage the time information in snapshot by sequential order and the spatial information in snapshot by AVL-tree. In addition, it presents the corresponding algorithms which improve the efficiency of topological operators.Thirdly, it provides an algorithm NWNDist for distance computation in transportation networks. The structure of NWN can efficiently store and access the underlying information in road network such as edges and nodes. It can also guarantee that all the close nodes are stored in the same unit (i.e., the same page) so that the time and numbers of accessing disks can be relatively reduced. In addition, R-tree is utilized to improve the efficiency of querying nodes. The algorithm can efficiently support various queries based on network distance, such as the shortest path, distance join.Fourthly, focusing on the nearest neighbor queries in road network, it presents the storage structure for network connectivity information and adds it in the index structure MON-tree to make it support the nearest neighbor queries. Then, a circle algorithm using the presented storage structure and MON-tree is proposed to support nearest neighbor queries in road network.