Design And Implementation Of Indoor Positioning System Based On WiFi

The location services has always been the necessary services in people’s daily life. With the rapid development of information technology and network technology, the location services can provide people with more and more accurate and efficient position information. The indoor positioning technology is one of the location services which has developed most rapidly in recent years. Besides, WiFi wireless network covers the metropolis at high speed in latest years, indoor positioning technology based on WiFi is given more and more attention by people and studied by more and more scientists. WiFi-based indoor positioning system has many features, such as low cost, high efficiency, strong utility. At the meanwhile, some factors,like weak filtration ability for the abnormal reference points and large of calculation in the Online phase, exist on traditional fingerprint positioning algorithms, these factors have bad effect on the positioning function.This paper studies indoor positioning technology based on WiFi and makes some improvement on the traditional fingerprint positioning algorithm, designs and implements a indoor positioning system based on WiFi.First, we give an brief introduction on the background of indoor positioning and the present situation of indoor positioning studied by scholars at home and abroad. we introduce and analyze the advantages and disadvantages of several common indoor positioning technologies, as well as the theories and features of indoor positioning methods based on distance and non-distance applied in indoor positioning technologies. In this paper, we choose the indoor positioning technology based on WiFi.Next, we lucubrate the fingerprint positioning algorithms which are applied on the indoor positioning technology based on WiFi commonly and give detailed introduction on the key factors which effects the functions of traditional fingerprint positioning algorithms and give the relative solutions, as well as introduces Affinity Propagation Clustering algorithms applied in this paper, As for clustering the reference points irregularly existing in traditional AP algorithm which has bad effect on the accurate positioning, We put forward a program to improve it.This program combines with positioning environment, in the Offline phase, the improved AP algorithm is applied to cluster the reference points, at the same time, the RSSI of reference points and physical positions are considered, and then the clustering result of the fingerprint database will be more reasonable. In the Online phase, first, we will simply determine the best positioning clusters, and then we can do the accurate positioning job by applying KNN within the best clusters, elvulate the positions of the targets.We have simulation test by MATLAB, and calculate the best positioning parameters by applying improved positioning algorithm AP-Knn in the test. After comparing and analyzing the simulation results of traditional algorithm and AP-Knn, we can show the advantages of the improved algorithm.And then, we make the demand analysis of the WiFi-based indoor positioning system, clear the system architecture and analogue functions, make detailed design on the Client and Server of the system and database, realize the functions of WiFi-based indoor positioning system by applying C#,MATLAB hybrid Programming and Android Programming.Finally, we create the experiment environment to test the functions of the system, and give detailed introduction on the positioning property of indoor positioning system which applies the AP-Knn algorithm through the experiment data, and verifie that the feasibility of the poisoning system in the large indoor place and a certain practical significance.