Dynamic nearest neighbour inquiry method on basis of regional coverage

Abstract

The invention relates to a dynamic nearest neighbour inquiry method on the basis of regional coverage and belongs to the technical field of mobile data index. The method comprises the following steps of: inputting position information of personnel; dividing a battle space into grids; indexing soldiers; recognizing the grids by using a Voronoi graph structure for regions with sparse soldiers in a battlefield; setting a value between 0 and 1 as a threshold value; when the density of the grids is more than the threshold value, determining the position of a nearest neighbour teammate by adopting a best-first nearest neighbour inquiry method (BNFF); and when the density of the grids is less than the threshold value, determining the position of the nearest neighbour teammate by adopting a Voronoi graph nearest neighbour method. By the dynamic nearest neighbour inquiry method on the basis of regional coverage, the cost of updating indexes is reduced. Meanwhile, compared with the existing method, the inquiry efficiency of the dynamic nearest neighbour inquiry method on the basis of regional coverage is obviously improved. The method can play a great role on the battlefield and provide a furthest rapid and accurate result for each requester.

Description

technical field [0001] The invention belongs to the technical field of mobile data indexing, in particular to a dynamic nearest neighbor query method based on area coverage. Background technique [0002] With the development of wireless communication technology, Global Positioning System (Global Positioning System, GPS) technology and database technology, mobile object database technology has gradually become a research hotspot in the database field, and the nearest neighbor query method of mobile object data is an important research topic. field, but most of the past research work is aimed at the nearest neighbor query in a static environment, and it has been relatively mature; for the nearest neighbor query problem in a dynamic environment where both the query object and the queried target are moving, At present, the research in this area at home and abroad is still in the initial stage; [0003] Although it is now in a time of peace, the research on war has been going on...

AI Technical Summary

Problems solved by technology

[0007] Su Chen and Beng Chin Ooi et al based on ST 2 Nearest Neighbor Query Method of B-tree Index Structure, ST 2 B-tree is a self-adjustable moving object B + Tree index structure, first in time, ST 2 The B tree will be the underlying B + The tree is divided into two subtrees BT 0 and BT 1 , each moving object selects the corresponding subtree to index according to the update time, and the two subtrees manage the moving object alternately with time; in space, according to the distribution of moving objects in the area, a reference point is selected according to the density set, and divide the area into blocks according to the reference point Voronoi block, and then select different grid granularity in different blocks to index the moving objects. For ST-based 2 The nearest neighbor query method of the B-tree adopts the method of continuously expanding the scope of the query radius until the number of moving objects in the query area meets the conditions of the nearest neighbor query. The disadvantage of the method is that the range query algorithm is continuously called, and the time cost is relatively large.

Images