Spatial-temporal data double-index structure based on LR tree and B + tree

Abstract

The invention provides a spatial-temporal data double-index structure based on an LR tree and a B + tree. The structure comprises an LR tree index structure and a temporal B + tree index structure; the layers of the LR tree index structure are tree root-node-leaf from large to small; a leaf node record item structure in the LR tree is defined as follows: (MBR, OID, MAP21Key), wherein the MBR is aminimum bounding rectangle containing an identification of an OID object; OID serves as object identification, the MAP21Key is an MAP21 conversion key value of an object life cycle LS, the MAP21 conversion key value is 0-dimensional point data formed by mapping a one-dimensional time interval of an object life cycle LS, the record items in the leaf nodes are linearly sorted and stored according toan MAP21 conversion key value, namely the size of point data; the hierarchy of the tense B + tree index structure is tree root-node-leaf in sequence from large to small, and the leaf node record itemstructure of the tense B + tree also comprises the MAP21 conversion key value. The spatial-temporal data double-index structure provided by the invention not only does not reduce the efficiency of time or space unilateral query, but also can consider the speed of spatial-temporal joint query.

Description

technical field [0001] The present invention relates to the technical field of spatio-temporal data indexing, in particular to a method based on LR tree and B + Spatio-temporal data double index structure of the tree. Background technique [0002] An important application in spatiotemporal information systems is spatiotemporal query, which can be divided into four situations: time point query, time segment query, time point based spatial query and time segment based spatial query. To solve the query problems in the first four cases, the key is to implement efficient spatio-temporal indexing in the system. In the existing research results, several effective multi-dimensional spatial indexing methods have been proposed, such as R tree, R + Trees, k-d trees, k-d-B trees, Grid structures, and improved versions of these methods, but none of them considered temporal query efficiency. On the other hand, many indexing methods based on effective time or transaction time have been ...

AI Technical Summary

Problems solved by technology

[0005] Aiming at the technical problem that although the existing unified index structure has more or less improved the speed of space-time joint query, it sacrifices the efficiency of unilateral query for space or time, the present invention provides a method based on LR tree and B + The double index structure of spatiotemporal data of the tree, the dual index structure does not reduce the efficiency of unilateral query of time or space, and at the same time can take into account the speed of joint query of space and time

Images