Rapid index method based on space-time data

Abstract

The present invention provides a wireless communication computer field, in particular relating to a spatial-temporal data fast index method, it is a high effect indexing mechanism, storing different time stamp space-time state using overlapping indexed structure, if the adjacent time slices have identical R-tree index branch, only one version is kept, this reduces the storage space. At the same time, data on one time slice adopts blocking index mechanism based on the four-fork tree no longer finding whole index space, the whole space is divided into n=sumd-1i=0(2k)i sub-index spaces (d is depth of the four-fork tree, k is dimensionality of the index space), only need find one subspace therein, greatly reduces the index time. Utilizing the index method is used for high efficient location of multiple wireless move node and data monitor management.

Description

technical field The invention belongs to the field of wireless communication computers, and in particular relates to a fast indexing method for spatio-temporal data. Background technique With the continuous development of modern communication technology, especially the diversification of wireless communication and the rapid increase of transmission rate, it occupies an increasingly important position in modern production and life. It is a very important task to establish an efficient spatio-temporal data index mechanism to quickly search and quickly locate a large amount of wireless data. Now large-scale commercial spatial database systems, such as Oracle Spatial, IBM DB2 Spatial Extender and IBM Informix Spatial Data Blade, all use R-tree index technology, as shown in Figure 1, which is the plane diagram of R-tree and the corresponding structural diagram. Its leaf nodes store the outer enclosing rectangle and spatial element ID of the corresponding spatial element. Non-l...

AI Technical Summary

Problems solved by technology

It can be seen from the structure of the R-tree that the search efficiency of the R-tree can be improved by allowing the spatial elements that are spatially close to each other to have the closest common ancestor as possible; however, the search of the R-tree starts from the entire index space. Searching in a full R-tree with a large depth, when the spatial dimension is high or there are many index objects, the spatial retrieval performance will drop sharply due to the increase in the overlap of intermediate node directory rectangles or repeated storage and the increase in tree depth ; and difficult to support time-varying spatial data

Images