Research On Localization Algorithm For Wireless Sensor Networks Based On RSSI

Wireless sensor network(WSN)is widely used in many aspects of our lives,such as military reconnaissance,hospital detection,smart home,environmental awareness and so on,thanks to its low energy consumption,low cost,strong adaptability,communication,storage and even mobility.In the research of WSN,there are many important technologies including node positioning technology,wireless communication technology,network protocol technology,time synchronization technology and network security technology.Among them,node location technology is the most critical link.In many application scenarios of WSN,if the correct node location information can not be obtained,the data obtained by the user will be meaningless.Therefore,designing a positioning algorithm with high positioning accuracy and high robustness has always been a research hotspot in the field of wireless sensor network.The research of this paper is mainly to improve the accuracy of the positioning algorithm of wireless sensor networks.The research content is as follows:First,research the theoretical knowledge of wireless sensor network and its positioning algorithm,and elaborate on the relevant content based on the existing literature.Second,different from the traditional trilateral positioning algorithm which adopts the anchor node selection strategy of selecting the three anchor nodes closest to the unknown node,this paper proposes an improved weighted trilateral positioning algorithm based on minimum uncertainty.The algorithm combines the minimum uncertainty and the triangle angle formed by the anchor nodes as the anchor node selection strategy.The selection criterion is not to directly select the three anchor nodes closest to the unknown node,but to measure the distance from the unknown node to each anchor node n times,then calculate the mean and standard deviation of the n distances,and select four anchor nodes with the smallest product of the mean and standard deviation.Then,the triangle composed of these four anchor nodes is tested for triangle interior points,and the coordinates of the anchor node groups satisfying the triangle interior point test are obtained by trilateral algorithm.Finally,the final result is obtained by weighting the previously obtained coordinates according to the criterion of “the closer the triangle formed by the anchor nodes is to an equilateral triangle,the lower the positioning error”.Simulation experiments show that the positioning algorithm has improved positioning accuracy under the same simulation conditions.Third,the classical centroid localization algorithm uses the centroid of triangle formed by the three anchor nodes closest to the unknown node as the estimated coordinates of the unknown node.Because the localization area is too large,the positioning accuracy is not high.Aiming at this shortcoming,this paper proposes an improved centroid localization algorithm based on order sequence,which combines the received signal strength region segmentation and the triangular centroid algorithm to effectively reduce the final localization area.Firstly,the triangle area obtained by centroid localization algorithm is divided by the perpendicular bisector of the connecting line segments of the three anchor nodes closest to the unknown node,and the order list of sub-areas is established.Then,the order sequence of the unknown node is obtained according to the RSSI value received by the unknown node,and the sub-area where the unknown node is located is determined by comparing the correlation degree.At the same time,another location area where unknown node is located is obtained through the triangular centroid algorithm.Finally,the two positioning areas are calculated as an overlapping area,and the centroid of the overlapping area is used as the final positioning result.The simulation results show that the positioning algorithm has been improved in positioning accuracy under the same simulation conditions.