Smartphone Based Indoor Localization For Symbolic Position

With the widespread popularity of the Internet and the rapid development of mobile intelligent devices, location-based applications and services are more and more closely related to people’s daily lives. While GPS cannot locate users indoor, there exist urgent need for new and more effective indoor positioning technologies. The applications not only need the positioning technologies to be accurate but also need them to be reliable, available and cost effective. Moreover, the indoor environment is complicated and of-ten lack infrastructure. Making the traditional indoor positioning technologies can not meet the requirements. With the need for better technologies, our research idea is to take advantage of the rich resources of smartphone sensing and computing in the indoor environment. Do not rely on any additional infrastructure or technical staff, making the positioning techniques to be accurate and at the same time improving reliability, availability and reduce the cost. This thesis proposed a symbolic indoor positioning technology framework, then proposed some solutions for the problems of indoor floor localization and metro line localization. We also proposed a common smartphone sen-sor calibration solution. At last, we designed and implemented the symbolic indoor positioning technology platform and developed a "Bit Campus" application. In sum-mary, this thesis has the following innovative contributions:● Proposed an infrastructure free indoor symbolic localization framework based on the smartphones (SIFIS). Traditional indoor localization technologies often rely on the infrastructure. We proposed a new indoor symbolic localization frame-work which make use of the technologies of activity recognition, crowdsourcing and pattern recognition. The framework is infrastructure free, ensure accuracy while improving reliability, availability and reducing the cost.● Based on the SIFIS technical framework. We proposed a solution for floor lo-calization problem, using only the smartphone acceleration, pressure and mag-netic sensors. We proposed the activity recognition algorithms based on ma-chine learning, the encounter detection algorithms based on trace matching and the symbolic positioning algorithms based on Dynamic Time Warping (DTW). Achieved the problem of infrastructure free indoor floor localization. The simu-lation and field study show that our approach can not only satisfy the positioning accuracy, but also has better reliability, availability, and lower cost.● Based on the SIFIS technical framework. We proposed a solution for metro train localization problem, using only the smartphone acceleration, pressure and mag-netic sensors. We proposed the train stop and door open detection algorithms based on state machines, the symbolic positioning algorithms based on Dynamic Time Warping (DTW). This approach only use the smartphone sensors for local-ization. Field study shows that our approach can not only satisfy the positioning accuracy, but also has better reliability, availability, and lower cost.● The smartphone sensor readings are not accurate before calibration, which will affect the accuracy of indoor localization. We presented a common sensor cal-ibration solution. With the help of crowdsourcing, it proposed the calibration algorithms for two sensors based on encounter detection and the calibration al-gorithm for all sensors based on shortest spanning tree. The calibration process is transparent to the users. It laid the foundation of data accuracy for indoor sym-bolic localization. The experiments to calibrate the barometer sensors show that our approach has good practicality and versatility.● Preliminary designed and implemented the indoor symbolic positioning technol-ogy platform. Developed a location-based social networking application. The application is based on our SIFIS platform, it describes the characteristics of a user by his location and recommend him new friends by the characteristics. The application shows the feasibility and advancement of our indoor symbolic local-ization technologies.