Parallel spatial object search with Optimal bayesian nearest neighbor in Grid regions

Abstract

Spatial database comprises of database that is optimized to store and newlinequery the data that is in a way related to objects involved in space including newlinepoints lines and polygons Database systems significantly use indexes but the newlineindex data is not optimal for using spatial queries Henceforth spatial newlinedatabases use spatial index to minimize the complexity arising due to the newlinespeed and provide solution for it by providing spatial index that includes newlinedatabase operations Spatial database facilitates the storage of spatial and nonspatial newlineinformation efficiently Such databases are highly required in the newlinecertain areas including environmental monitoring space urban planning newlineresource management and geographic information system newlineHowever identification of the Nearest Neighbor NN object search is newlinea vital part of spatial database But they had some difficulties while newlineperforming NN in uncertain spatial database as it does not support high newlinedimensional data structure which includes high communication overhead newlineRecent works carried out considers Bayesian Nearest Neighbor Search newline BNN in a spatial database which provides an efficient method to identify the newlinenearest neighbor on spatial database without any ambiguity newlineThe work starts in a direction of identifying the nearest neighbor newlineobjects efficiently Bayesian Nearest Neighbor for NN search and similarity newlinesearch in an uncertain spatial database not only achieves NN search newlinepowerfully and recovers the distance information not only from single server newlinebut also from distributed servers BNN performs NN search and similarity newlinesearch are applied to the high dimensional data structure which results newline newline