Desing And Implementation Of PDR Aided Positioning SDK Based On The Adnroid Platform

In recent years,mobile intelligence has developed rapidly with the rise of Location Based Services.Thus,an increasing number of mobile developers participate in the development of LBS applications.At the same time,this situation puts forward higher requirements on the accuracy,stability and energy saving of location information acquisition.According to the above requirements,we design and implement a SDK based on Pedestrian Dead Reckoning(PDR)which can reduce energy consumption and improve the positioning accuracy under the Android platform.Firstly,a framework that can accommodate and expand different positioning,location and fusion algorithms is designed.The following algorithms are packaged in this framework:the GPS,base station,Wi-Fi positioning algorithm,PDR algorithm for the localization of non-oriented smartphone,and the fusion algorithm based on particle filter.Secondly,a series of auxiliary tools is implemented to expand the function of SDK.It contains the map tool for displaying the positioning results and the description of the motion trajectory,and the chart tools for analyzing the positioning data.Finally,a unit test is run to verify the feasibility of the functions mentioned above.And then,we using the unconstrained PDR algorithm on smartphone to assist the Wi-Fi fingerprint location based on the particle filter fusion algorithm.Compared with using Wi-Fi fingerprint-based localization alone,we verify the feasibility of the fusion localization algorithm.To sum up,this thesis provides a feasible location SDK for LBS application developers,and provides a convenient research environment for PDR algorithm researchers.