Design And Implementation Of Indoor Positioning System Based On Android

With the development of mobile Internet technology and the increasing demand of location-based services(LBS), indoor positioning technology has been developed rapidly. A wide range of indoor positioning technology based on mobile terminals is studied extensively. For example, the WiFi positioning technology, bluetooth positioning technology, image based positioning technology and inertial navigation technology, etc. The inertial navigation technology is not dependent on external signals, which has the advantages of high stability and low cost. WiFi positioning technology is widely used because of the wide spread of AP, especially WiFi fingerprint positioning technology, which has high accuracy, simple and low cost advantages.This paper focuses on the pedestrian inertial navigation technology, WiFi fingerprint positioning technology and machine learning positioning technology, deeply studies its accuracy, stability and complexity. The advanced algorithms and methods are considered in detailed.Firstly, the principle of existing inertial navigation technology and the main reason affecting the positioning accuracy are analyzed. The performance of existing pedestrian inertial navigation method needs to be optimized in gait detection and particle filter aspects. Therefore, a dynamic power threshold adjustment method for gait detection is proposed. At the same time, the map information is added to improve the particle filter algorithm, which improves the accuracy of the pedestrian navigation. Finally, the proposed method is implemented on Android platform, and tested through Android phone. Experimental results show that the proposed method is more accurate than the traditional method.Further, the main problem existing in the practical application of the pedestrian inertial navigation is analyzed. To solve this problem, a WiFi fingerprint aided pedestrian inertial navigation technology is proposed, the position of the pedestrian inertial navigation system is corrected by the proposed algorithm. The proposed method is implemented and tested on the Android platform. Experimental results show that the proposed technique can solve the problem of pedestrian inertial navigation error effectively, and has high precision and stability.Finally, the accuracy and computation time of machine learning algorithms in indoor positioning are studied. Then a machine learning algorithm with high positioning accuracy is chosen, and a positioning technology based on the combination of machine learning algorithm that is selected and pedestrian inertial navigation is proposed. The proposed method is implemented and tested on the Android platform. Experimental results show that the method has a higher positioning accuracy compared to the fingerprint positioning method and the positioning technique mentioned in the last chapter.