Research On Indoor Positioning Algorithm With Smartphone Based On EKF And PDR

In recent years,the rapid development of Internet and mobile communication technologies has led to the widespread application of Location-based Services(LBS)in many fields of livelihood.Data of location includes three important aspects: time,location,person.GPS,Bei Dou and other technologies have met the needs of outdoor positioning.However,these technologies cannot have an effective influence on the indoor area due to the shelter of cement buildings.There are many disadvantages in existing indoor positioning method.For example,some indoor positioning system operated many expensive hardware which caused the difficulty in popularizing,and some used hardware with high power consumption.This thesis proposes a new positioning system based on the existing indoor positioning.This system relied on Bluetooth module and IMU in smartphone and considered both low cost and low power consumption.This thesis focuses on the following aspects:Firstly,this thesis studied how to accurately measure the steps using the IMU in the smartphone by analyzing the characteristics of human walking.This thesis proposes a window filtering method for three-axis accelerometer,this method can improve the accuracy of steps measurement.Secondly,this thesis studied how to dead reckoning by using smartphone.The process includes step number measurement,step length measurement,direction measurement,position calculation,etc.This thesis studied several methods of step length measurement that commonly used,and finally selected dynamic step length measurement method.In addition,this thesis also studied how to measure the direction by using magnetometer,and then achieved the PDR.Thirdly,this thesis proposed a new method to correct ‘drift' of PDR using Extended Kalman filter.The walking coordinate sequence of PDR as an observation model in this method,and this sequence included some coordinate of Bluetooth which user close to when they walking.And then user can get pure location information after using EKF with this sequence.Finally,this thesis proposed a new method which is Bluetooth Cooperation near-neighbors.This method took advantage of the characteristics of Bluetooth two-way communication.The thesis also proposed two kinds of data organization when the Bluetooth broadcast,Mode A and Mode B.Mode A was used for indoor positioning system with smartphone and server;Mode B was used for indoor positioning system that only with smartphone.