Research On The Location Of Mobile Resources Based On Demand Density Topology Perception

In view of the publicity and service characteristics of movable resources,we must consider not only the demand density of the area,but also the scope of services when we make location decisions for mobile resources,so as to maximize the use of limited resources to meet the service needs of users.Aiming at the problem of mobile resource location,this paper proposes a mobile resource location method based on demand density topology awareness.The main research work is as follows:(1)We propose a method for filtering resource demand density.It aims to select the site of movable resources,using limited resources to maximize the needs of users,and improving economic and social benefits.First,we grid the site selection area,count the population distribution in each grid,and set a threshold to get the resource demand distribution points in the area.Second,we use the Delaunay triangulation method to surface the resource demand distribution points in the area,calculate the distance between each distribution point and its neighboring points,and then remove the sparse points according to the distance distribution.For removing the distribution points in the sparse area,we use the distribution characteristics of the distance between the distribution points.First calculate the connection distance between each distribution point and its neighboring connection points,then calculate the average connection distance of each distribution point,and finally remove the distribution points with a larger average distance according to the distribution characteristics of the average distance.The remaining demand density distribution points are basically in demand Dense area.(2)We propose a perception algorithm based on self-organizing neural network.The self-organizing neural network algorithm is used to perform topological perception of the filtered resource demand distribution points.The number of neurons in the self-organizing neural network algorithm can be controlled,which can solve the problem of maximizing the resource demand of users with limited resources.(3)We propose a method to homogenize the perception points.The demand distribution points perceived by the topology are not uniform,so a method for homogenizing the sensing points is proposed.First,we use the K nearest neighbor algorithm to find the Euclidean distance between each point in the distribution point and the K nearest neighbor points around it,and calculate the average distance.Then we draw the frequency distribution histogram of the average distance and judge its distribution characteristics.Finally,we find areas with higher density,reduce the number of demand density distribution points,and try to make the sensing points uniform.(4)We conduct experiments and analysis on the method in this paper.The first part is the simulation experiment,which are the experimental verification of demand density filtering method,the experimental verification of demand perception and perceptual point homogenization,and the application of data filtering in sampling consistency.The second part is an example analysis.Three data sets are used to verify the method in this paper.Then,the basis analysis method is used to experiment on the three data sets.Finally,the method in this paper is compared with K-Means and GMM algorithms and fuzzy multi-objective immune genetic algorithm.