Wi-Fi Indoor Location Method Based On The Collaboration Of Assistant Node

With increasing development of mobile Internet technology over the world,Wi-Fi network becomes more and more popular in daily life,playing a more important role in the society than before.Furthermore,multiply services based on location attract people.Because of low cost,high implementation,high-precision positioning,good scalability and other advantages,many researchers focus on RSS location algorithm based on Wi-Fi so this topic has quickly become a hot spot in the field of current indoor positioning technology.This paper makes a summary of the current research status and development on indoor location algorithm based on Wi-Fi at home and abroad.Based on the similarity of RSS time series,assistant node is selected,and the NLOS error elimination algorithm with modified Kalman filter is applied into TOF ranging measurement between nodes.Meanwhile,the Exhaustive method and the Maximum Gradient Descent method are used to optimize the positioning error.According to the analysis of traditional performance evaluation based on RSS or TOF localization algorithm,the influence of the number of assistant node and access points respectively on the performance of the algorithm is analyzed.The main work of this paper is as follows:1.Assistant nodes based on the similarity of RSS time serials.According to the past literature,RSS-based on Wi-Fi indoor positioning technology,the localization accuracy of traditional fingerprinting algorithm mainly depend on the accuracy of matching algorithm and fingerprint database.And it is generally accepted that when considering the information of a single node,the information of other nodes can not be fully effectively taken in account in the indoor environment.In fact,there are so many Wi-Fi devices in the current indoor environment and the information of these Wi-Fi devices can be used to improve the positioning accuracy.Therefore,appropriate assistant nodes based on the similarity of RSS time serials are elaborately selected around the unknown nodes,which would be applied to enhance the localization accuracy.2.The Wi-Fi indoor localization method based on collaboration of assistant nodes.Firstly,appropriate assistant nodes based on the similarity of RSS time serials are elaborately selected around the unknown nodes and then distances between them are used as auxiliary information to improve the positioning accuracy.However,the ranging error of time-of-flight measurement is not subject to Gaussian.Meanwhile,an adaptive Kalman filter with colored noise is used to mitigate the time-of-flight ranging error in the complex indoor circumstance.After that,the searching model would be built based on the range between every two nodes.Finally,the Exhaustion and the Maximum Gradient Descent method would be used to optimize the positioning error.Experiments demonstrate that in complex indoor environments,our system can outperform its counterparts with more robust performance and lower localization estimation error.3.Cramer-Rao Lower Bound Analysis of Wi-Fi Indoor Localization Using Fingerprint and Assistant Nodes.The current research status of the performance evaluation based on RSS or TOF localization method is analyzed in this paper.And the influence of the assistant node and the number of access points on Cramer-Rao Lower Bound would be deeply analyzed and the experiments have demonstrates its influence on the Wi-Fi indoor positioning method based on the cooperation of assistant node.