Space-time index establishing method based on frequency attributes and PCA

Abstract

The invention discloses a space-time index establishment method based on frequency attributes and PCA. The method comprises the following steps of: 1) data acquisition: acquiring and acquiring enoughdata with four characteristics (called frequency attribute-based space-time characteristics), namely longitude, latitude, time and query frequency generated by an object in a real application scene, wherein the four characteristic fields of the data cannot have any loss; 2) data preprocessing: regarding all the acquired data in the currently divided data block as points in a space-time three-dimensional coordinate system, selecting first K (default to be 4) data points with the maximum frequency, and obtaining projection planes corresponding to the four points according to a PCA algorithm; wherein the four points are connected to form a convex quadrangle according to the four points on the projection plane, the midpoints of the four edges are selected, and the midpoints of the opposite edges are connected to construct two division planes, and 4, index establishment, recursion of the division operation in the step 2 and the step 3 are carried out, and each divided space block can correspond to one node in the tree.

Description

technical field [0001] The invention relates to the field of data index establishment, is mainly used in databases, and in particular relates to index establishment of spatio-temporal data based on frequency attributes. Background technique [0002] With the rapid development of sensor technology, the spatial positioning technology and communication technology based on the Internet of Things, cloud computing, and mobile computing are becoming increasingly mature, and a large amount of data with spatiotemporal information generated by the activities and interactions of objects in various application scenarios can be detected, Transmission and recording. In the face of a large amount of spatiotemporal data, how to store the spatiotemporal characteristics of these data is facing a huge challenge. [0003] When processing large-scale spatio-temporal data, we often encounter the problem of low query efficiency of spatio-temporal data. Large-scale data must be stored in the data...

AI Technical Summary

Problems solved by technology

[0003] When processing large-scale spatio-temporal data, we often encounter the problem of low efficiency of spatio-temporal data query

Large-scale data must be stored in the database, and the most important aspect of the benefits brought by the database is the significant improvement of query efficiency. However, in the face of data with temporal and spatial characteristics of frequency attributes, the index established by traditional relational databases for data temporal and spatial characteristics It often cannot bring about a significant improvement in the efficiency of spatio-temporal query, and it is necessary to adopt targeted methods to build indexes according to the characteristics of data access frequency to cope with the increasing demand for spatio-temporal query

Images