Indexing And Querying Of Moving Objects Based On RQN Tree In Networks

Moving object database technology is the new research area of computer science. It is used mainly in managing the moving objects which are changed continuously by time. It has extensive application demand. As the computer network and wireless location technology developing rapidly, the mobile information service has been getting attention. The index technology of the moving object is borned to effectively manage a great number of data of moving object database accumulated steadily by time. It is not only able to keep the update performance of the index, but also manage the moving object effictively so as to meet the needs of actual application. Moving object database technology is the key point in moving object database area.Many people think that Moving objects should move in Euclidean free space, but in fact, the movement of moving objects is usually limited in the topology network (road-network), such as the car driving in the road network. Through the research of the existed moving object database technology and the application demand of the restricted moving objects, it analyses the principle which should been persisted in the building of the road-network index and puts forward RQN tree which is more suitable to index network. Through the R tree forests and the HASH structure, it proposes an index method of Moving objects based on RQN tree.RQN tree divides the space of the road-network into multi-layer sub-index space according to the road distribution. The Leave nodes in the bottom contain many sections and each section corresponds to an R tree. The R tree establishes an index to the moving objects in corresponding section according to the time segment.as the steady accumulation of time, It inserts the record of time segments to R tree actively.According to the time sequence, the HASH structure connects every moving objecte's track divided by time and space.It consists of the hash table and the dynamic link list. There are the identifiers of the Moving objects in the hash table.Each node points to a dynamic link list. There is the address information of the Moving objects track kept by time sequence in the node of the link list.In this paper, we use the T Brinkhoffs moving object simulator and real road traffic map to produce the experimental data of moving objects on the road.We compare the performance of dynamic combined indexing with an RQN tree indexing by the number of node access. The result shows, the RQN tree indexing has edge on road search and window search.