Research On Indoor Positioning Technology Based On Android With Fusion Of PDR And WiFi Fingerprint

With the help of the technical power of three major communications operators 4G + and rapid development of relevant industries based on the "Internet +", LBS has been widely applied in marketing, disaster relief, warehousing logistics and many other fields. In outdoor environment, the location services, based on the GPS or BDS, are thoroughly studied and widely used. In indoor environment, BDS and GPS fail to provide location services, and there are many flaws in the mainstream indoor positioning technologies. Based on IMU, the data acquisition sensors of dead reckoning positioning technology are separate and fixed independently in some parts of the moving objects in order to achieve ingenious collection of motion data and design algorithms, which limits the versatility of this technique. This paper explores the positioning characteristic of IMU and WiFi fingerprint location technology, and studies indoor positioning technology based on Android with fusion of PDR and WiFi fingerprint. The research emphases are as follows:(1)Improving pedestrian pace recognition algorithms. With the help of the analysis of the change rules of acceleration sensor signal caused by the pedestrian pace and based on local variance analysis and detection methods, the new method adds the constraints of pace's starting and ending time point which are approximately symmetry in order to introduce the pace recognition algorithm based on variance domain. Comparing with the local variance analysis detection method, the new pace recognition method proposed in this paper further regulates the factors necessary for detecting steps, improves the accuracy of step detection, and also helps to define time points for the heading estimation.(2)Proposing micro-heading angle estimation methods based on micro-scene. In indoor environment, the direction sensor and gyroscope can be disturbed by the external factors, which leads to relatively large errors of heading angle estimation. Therefore, this paper proposes micro-heading angle estimation methods based on micro-scene. In the micro-scene, micro-heading angle measured by two sensors can provide corroboration to another one. It improves the accuracy of the measurement of micro-heading angle so as to improve estimation accuracy of whole pace heading angle.(3)Elaborating integrated indoor positioning methods based on Android. Aiming advantages and disadvantages of PDR positioning technology and WiFi fingerprint positioning technology, the paper integrates two kinds of positioning technologies on Android. WiFi fingerprinting technique eliminates cumulative errors in the process of PDR positioning,and PDR positioning technology refines Wi Fi fingerprint positioning accuracy. By means of theoretical analysis, Wi Fi fingerprint and PDR positioning technology have complementary advantages, and helps to improve positioning accuracy.Finally, experiments show that the pace recognition algorithms and heading angle estimation method studied in this paper further improve the integration of indoor positioning accuracy which can better meet customers' demands.