Research On Indoor Positioning Algorithm Based On Inertial Sensors And Wi-Fi Signals

Location Based Services (Location Based Service, LBS) have been extensively studied for the past few years, and it plays a very important role in people’s life. The outdoor localization is mature while indoor localization needs new techniques because the GPS signal in indoor environment is too weak to accurately offer positions. In this thesis, combining existing work and our own working experiences, we develop a reliable algorithm called Inertial Sensors and Wi-Fi Signal Based Algorithm (ISW). This algorithm is divided into step detection and orientation estimation solving calculating step length and walking direction problems.For the detection of steps we first use build-in 3D accelerometer sensors to set up step boundary module. We collect walking time, acceleration characteristics to mark step boundary. Then, we analyze the Step Mapping Module and utilize Wi-Fi signals to automatically calculate the step length. The feedback results will be given to the Step Mapping Module so that the drift errors could be reduced when the user is walking.For the orientation of pedestrians we propose a method using magnetic field sensor and the additional integration of gyroscope. We analyze the characteristics of these two sensors and changes of data to reduce the effect of magnetic interference in case of calculating the walking direction accurately. These two sensors can be combined to almost get real orientation and enhance the location accuracy. Finally, we combine these three results to estimate locations and draw pedestrian lines on the floor map.In this thesis, we use Android based smart-phones as the hardware platform to design and implement this algorithm. At the same time, we also implement three popular algorithms: DTW, Zee and FSM. We debug in experimental environment and compare the results with other recent studies. The experiments show the improvement of tracing accuracy compared to other traditional indoor localization algorithm in indoor environments under the influence of nearby complex ferromagnetic interference. This algorithm has low costs and the workload is small. It is easy to popularize, having high precision, and has great significance.