Research And Implementation Of Multimode Indoor Localization Based On Smartphone

In modern life, with the increase of peoples’ indoor activities and the buildings’ area, a variety of services based indoor location have emerged, which require to obtain peoples’ accurate indoor location information. Although the GPS technology has been widely used in outdoor, it can’t be used indoors because the signals are shielded by the construction, and so indoor localization is still an unsolved problem in the world. This is mainly because that the complexity of the indoor environment is prone to cause multipath propagation phenomenon of signals, and most of the reinforced concrete structure would produce an electromagnetic signal shielding, in addition, placing of goods, peoples’ movement and on-off states of door and windows would affect the distribution of indoor signals. Although scholars from many countries have tried a variety of localization technologies, but there still doesn’t appear one kind of indoor localization solution which can be recognized by the public.This article fully references the results of previous studies, presents multimode indoor localization technology based on smartphone. This technology uses multimode replacement strategy to combine Wi-Fi fingerprint localization technology and PDR pedestrian dead reckoning localization technology. Based on the above rationales, a multimode indoor localization system is designed on the Android platform, and many problems of engineering details are overcome in concrete implementation.In this paper, the research content mainly has the following several aspects:(1)The platform features of Android, system framework and built-in sensors system are deeply studied. In the study of built-in sensors system in Android, we focus on the learning of application scenarios by using smartphone’s sensors, source code structure of built-in sensors system, calls method of built-in sensors system, Wi-Fi management and hotspots scanning. The study of this part provides the basis for the development of multimode indoor localization system based on smartphone in this paper.(2)The data space coordinate transformation algorithm with built-in accelerometer of smartphone is deeply studied. This paper attempts by quaternion algorithm and the direction cosine algorithm, finally finds an effective way to overcome the phenomenon that data collected by acceleration sensors would be constantly shifting due to the changes of smartphone’s posture, this ensures sensorsdata can accurately reflect actual motion state no matter in what gestures smartphone are. The study of this part provides the preconditions for the subsequent data collection.(3)The implementation process of Wi-Fi fingerprint localization technology and PDR pedestrian dead reckoning localization technology are thoroughly studied.Among them, in the study of Wi-Fi fingerprint localization technology, we focus on the detailed analysis of offline training stage and online localization stage, also studies the construction method of Wi-Fi fingerprint database; in the study of PDR pedestrian dead reckoning localization technology, we focus on the detailed analysis of the detection of pedestrians’ walking steps, the estimation of pedestrians’ step length and the estimation of pedestrians’ walking heading. In addition, a lot of works have been devoted to research on the multi-mode replacement strategy.In the localization experiment, the system proposed by this paper has been compared with traditional Wi-Fi fingerprint localization technology in the practical environment; test results show that this system has better localization effect.