Study On Indoor Positioning Technology Based On Smartphone

In recent years,with the rapid development of mobile Internet,the demand for location-based services is growing fast.Indoor positioning system uses short-range wireless technology and pedestrian dead reckoning,for example,WiFi,Bluetooth,Infrared,RFID,UWB and so on.With the popularity of smartphone,smartphone carries a variety of sensors,meanwhile,it has a lot of advantages,such as,lower cost,convenience.More and more scholars research indoor positioning base on smart phone.This paper studies the WiFi location fingerprint technology and pedestrian dead reckoning technology.This paper merges two methods to implement three-dimensional indoor positioning based smartphone.There are a lot of access points in WiFi positioning.Maximum mean AP selection algorithm(MM)chooses the AP based on the mean of signal strength.It can't show that there are different signal strengths of same AP from different samples.This paper proposes maximum cumulative difference(MCD)AP selection algorithm based on maximum mean AP selection algorithm.MCD algorithm calculates the difference of signal strength from different samples.It chooses the AP based on the cumulative difference of signal strength.Experiment shows that positioning accuracy of MCD algorithm is 6% higher than the MM algorithm,and has better stability.In the conventional PDR technology is mainly used to locate the two-dimensional space.We find that the step high is related with the amplitude and cumulative deviation of acceleration through analyzing the acceleration of upstairs and downstairs.We pose step high estimation model and prove its feasibility with experiment.This article implements three-dimensional positioning using step high estimation model and PDR technology.We analyze the advantages and disadvantages of WiFi location fingerprinting positioning and PDR positioning,and propose a fusion strategy based on WiFi fingerprint positioning and PDR.The final positioning result of co-locating is weighted by WiFi positioning result with PDR positioning result.Finally,we analyze experimental result based on WiFi fingerprint positioning and PDR.The result shows that the combination of WiFi and PDR provides higher accuracy positioning.