Design And Implementation Of Indoor Location Algorithm Based On RSSI

With the rapid development of wireless sensor networks,indoor positioning and navigation is widely used in various occasions,so it attracts a large number of researchers to conduct in-depth research.Among them,indoor positioning technology based on received signal strength indicator is widely concerned because of the great advantages in power consumption,transmission distance and hardware cost,which focused by researchers.Therefore,the indoor positioning technology based on RSSI has important research significance and practical value.The weighted centroid localization algorithm based on RSSI combines the advantages of RSSI ranging algorithm and geometric centroid method,which is simple and does not need additional hardware.However,the weighting factor of weighted centroid localization algorithm based on RSSI is vulnerable to dynamic environmental interference.Aiming at the problem,this thesis improved the weighted centroid localization algorithm,according to the characteristics of nearest neighbor classification based on K nearest neighbor method,different weights are assigned to different beacon nodes to avoid the irrationality of weight selection.First of all,the least squares method is used to fit the logarithm of the signal intensity and distance in the location of unknown node in this thesis.The distance d is used as the weighting factor;then,the K nearest neighbor method is used to assigned a reasonable weight correction factor to the weights of the beacon nodes;Finally,the location information of unknown nodes can be calculated more accurately by the improved weighted centroid localization algorithm.In the area of 225 square meters,the experimental results show that the average localization accuracy of weighted centroid localization algorithm is 2.03m and the improved algorithm is 1.44m.Compared with the original weighted centroid localization algorithm,the average positioning accuracy is improved by 29%,the positioning accuracy is improved by 38%greatly in the edge region.In this thesis,the indoor positioning algorithm meets the design requirements and indicators in both function and performance.which can meet most of the requirements of indoor localization.