Quadrilateral Weighted Centroid Algorithm Based On RSSI Range Correction

With the continuous development of wireless communication technology,wireless sensor network is also applied from the beginning of the military field to today's industry,medical,agriculture and other fields.Its network terminal is a large number of intelligent sensors.Human society perceives,collects,stores,processes and transmits the situation of local area through sensor nodes.Complete the monitoring of the physical world.In every field of wireless sensor network application,the location information of nodes is needed,so it is a hot topic to improve the positioning accuracy of the node.This paper mainly through two aspects to improve the positioning accuracy of the node,one is the range correction of the received signal strength indication,the other is to improve the positioning algorithm of the node to reduce the error.Wireless sensor networks are usually used in NLOS environment The wireless signal will be affected by multipath,diffraction and other environmental factors in the transmission process,resulting in large differences between the received signal strength indicators(RSSI)from the same node measured by unknown nodes in a short time.If the distance between RSSI value nodes with larger error is selected in the ranging model,then using this low precision distance to participate in unknown node localization will greatly reduce the positioning accuracy.In this paper,we will focus on the analysis of the Kalman filter algorithm.First,we introduce the principle of the Kalman filter,and then establish the corresponding Kalman filter model based on the RSSI experimental model,and then optimize and adjust the parameters of the Kalman filter through experiments to make it have better filtering effect,Then Kalman filter is carried out according to the RSSI values under different noise intensities.Finally,the Kalman filter is compared with other filtering algorithms.The experimental results show that the Kalman filter has better filtering effect under different noise intensities,and the Kalman filter has better convergence speed and smoothness under the same conditions.An improved quadrilateral weighted centroid algorithm based on least squares is proposed.In the traditional quadrilateral weighted centroid algorithm,a new method is proposed to solve the problem that the adjacent anchor node circles do not intersect,which avoids selecting the anchor node group with larger error to participate in the positioning and reduces the error accumulation.Then,the estimated position of the unknown node is obtained by the least square method,and the inner intersection of the four adjacent anchor node circles is selected by the estimated coordinates,The nodes with higher accuracy are retained to participate in the weighted centroid positioning,so as to improve the positioning accuracy.Finally,the improved quadrilateral positioning algorithm based on location assistance(OLS-QWCRC)is compared with the range correction triangle(TWCRC)and quadrilateral weighted centroid algorithm(QWC).The experimental results show that the OLS-QWCRC algorithm has better robustness and improves the positioning accuracy.