Location Method Of Smartphones In Subway Environment

With the development of economy,smart phones are becoming more and more popular in China.Based on powerful hardware,the navigation software on mobile platform is quite a lot.Smartphone uses GPS and GSM to achieve high positioning accuracy with fast response characteristic,which greatly facilitate our lives.But when people throw away traditional paper map,hugging convenient smartphone electronic map,there is still some places that smart phone navigation unable to cover,such as the subway.Since satellite signals can not penetrate the soil,the GPS location can not be used.Due to base station signal reflect in the underground tunnel,base station positioning error is large and obvious transition.So we can't effectively get positioning results by base station system.Currently the mobile phone positioning algorithm does not differentiate according to the above and under the ground,which we can get high positioning accuracy on the ground but low accuracy under the ground.In summary,the current mobile phone map software can not meet the requirements of position accuracy.There are two ways to improve the positioning accuracy of mobile phone navigation software in the subway.On the one hand,we can seek cooperation with subway companies to get metro scheduling information.On the other hand,we can use various sensors in smart phones to improve mobile phone positioning accuracy.However,subway companies can't share their information for the safety of metros.And the cost of tunnel renovation is also unrealistic.Therefore,to improve the positioning accuracy of mobile phone navigation under the ground,we can only use sensors to create a zero-cost mobile phone navigation system.Since the inertial sensor has not particular restriction in the environment,it can compensate the insufficient of GPS and base station location.And many smart phones are equipped with accelerometers and gyroscopes.So,inertial sensor can be used to improve the accuracy of underground positioning.The main work of this thesis is as follows:(1)We propose an inbound passengers judgment policy which is based on mobile phone base stations with CellID.This algorithm can efficiently and automatically implement switching between the traditional orientation and positioning of the subway.(2)In order to determine the relative attitude between the mobile phone and the train,we proposed a gradient descent based estimation algorithm to solve the initial moment of inertia positioned relative positional relationship between the phone and the train.We verify the algorithm effects and get good results.(3)We analyze error sources for zero drift during positioning system output speed.We found that the error comes from the gyro zero drift and accelerometer noise itself.According to the data fusion theory,we use the accelerometer which has high reliability of measurement for a long time to make up gyroscope measurement error.At the same time,we focused on the speed zero drift of the whole system and proposed an error correction algorithm based on reference point.The algorithm is verified by experiment.(4)For inbound detection system,the reference point detection and positioning error,we conducted a lot of experiments,draw a graph showing the cumulative distribution system error assess the overall positioning system performance.