Research On Centroid Localization Model Of WSN

As a relatively new technology at present,WSN technology is largely applied to military,medicine,disaster prevention and other fields,and widely used in fire monitoring,earthquake prevention and control,target tracking and other aspects.Node localization is one of the technologies of WSN.The localization algorithm is the kernel of the localization technology.The number of nodes,the distance of nodes and the environment during the process of node positioning,will affect the accuracy of node positioning.At present,many people have studied the node positioning technology,but most of the existing centroid positioning algorithms take the distance as the weight.When using RSSI for ranging,they ignore that the RSSI value is extremely vulnerable to interference,which leads to the inaccuracy of RSSI value and ultimately affects the positioning accuracy.At present,most of the researches only use the average filter or Gaussian filter to filter the RSSI value and the effect is not particularly obvious;in positioning,few people consider the impact of positioning error on positioning accuracy,just simply take the distance between nodes as the weight,which exist some problems in the selection of weight in weighted centroid positioning algorithm.Based on this,this article will research and improve the node positioning algorithm,combined with the advantages of RSSI ranging and centroid positioning algorithm,considering the impact of positioning error on positioning.The contents and innovations of this paper are as follows:1.In view of the situation that some RSSI values in RSSI ranging may be interfered and lead to inaccuracy,the improved hybrid filtering function is used to filter out the RSSI values with larger error,and the more accurate RSSI values are selected.2.After getting the filtered RSSI value,convert it to distance.On this basis,the coordinate positions ofC_n~3 estimated target nodes close to the real distance are obtained by trilateral measurement method for n anchor nodes.3.The k-means algorithm is simple to achieve,but it is easy to get local optimal solution.Therefore,the k-means++algorithm is introduced into the WSN positioning,and a weighted centroid location algorithm based on the improved k-means algorithm is proposed.The algorithm is used to cluster the aboveC_n~3 estimated target node coordinates values,and finally the density in the cluster is used as the weight to calculate the centroid to get the node estimates the location.Use MATLAB simulation tool to conduct a simulation experiment on the proposed RSSI ranging hybrid filtering algorithm and the improved centroid positioning algorithm.The experimental results prove that the improved filtering algorithm has a significant improvement effect on the RSSI value filtering,and can effectively filter out the inaccurate RSSI value.The improved centroid positioning algorithm has a effective ameliorate effect on the positioning accuracy compared with the existing positioning algorithm.