RSSI-based Localization Error Optimization Method For Wireless Sensor Network

As one of the key technologies of Wireless Sensor Networks(WSN),localization technology plays an important role in geographic localization acquisition,load balancing,and network routing.Most of the nodes scattered in the wireless sensor network are relatively inexpensive and cannot be located with additional equipment.If they want to achieve their own localization,they can only use the information obtained by communication between nodes to locate.For example,the node localization method based on RSSI ranging is to calculate the distance between the unknown node and the anchor node by establishing a signal transmission model,and then calculate the coordinates of the unknown node according to the distance.This method is simple and easy to implement,but it is susceptible to the surrounding environment such as multipath fading,shadow effect,etc.,which causes the signal to be unstable during transmission,causing errors in the distance between unknown nodes and anchor nodes,thus affecting the location of unknown nodes.error.In view of the above problems,this paper corrects the ranging error and the node localization error respectively:(1)In the ranging stage,this paper proposes optimization algorithm of RSSI transmission model for distance error correction(RSSI-DEC).The parameter path loss factor and the reference path loss between anchor nodes are optimized.Firstly,the fitness function of the algorithm is determined according to the principle of square error of the ranging error,and the logarithmic attenuation inertia weight function is proposed for the problem of premature intelligent optimization algorithm to balance.The algorithm's local optimization ability and global optimization ability,then use FA algorithm and PSO algorithm to optimize the model parameters,and obtain the model parameters of the monitoring environment to adapt to the node,to determine the signal transmission model that is more suitable for the monitoring environment of the node.Finally,the four-sided measurement method is used to realize the node localization in the three-dimensional environment.(2)In the node localization stage,this paper proposes an improved RSSI-DEC-based three-dimensional weighted centroid localization algorithm(WLRSSI-DEC),which first combines the principle of projected coordinates in three-dimensional space to interact with each other in the communication region of unknown nodes in three-dimensional space.The anchor nodes of the communication are arbitrarily selected four to form a positional relationship of four spheres with the unknown node as the center of the sphere and the distance between the anchor nodes and the anchor nodes as the radius,and the formed four-ball positional relationship is respectively projected onto the three coordinate planes.Then the calculation of the projected coordinates is carried out,and finally the weighting factor based on the power decision is introduced.This not only solves the influence of the distance between the nodes on the coordinates of the unknown nodes,but also prevents the weighting factors from over-correcting the coordinates of the unknown nodes.The simulation results show that the error optimization method proposed in this paper has good range error optimization effect and better environmental adaptability in the ranging process,and also achieves the purpose of optimizing the localization error in the node localization process.localization accuracy.The improved RSSI-DEC-based three-dimensional weighted centroid node localization algorithm proposed in this paper has an average relative localization error of about 9.01%,which is 22.46% lower than the RSSI-based weighted centroid localization algorithm(WRSSI),and the adaptive RSSI localization algorithm(ARSSI).The average relative error is reduced by 6.11%.Therefore,the RSSI-DEC algorithm proposed in this paper has a good ranging error optimization effect,and the WLRSSI-DEC localization algorithm localization error is significantly optimized.