Research On Pedestrian Indoor Location Based On PDR-WiFi Technology

In recent years,with the emergence of wireless communication technology and a large increase in mobile communication equipment,a variety of demand services,such as security rescue,shopping malls navigation,personnel tracking,smart home,etc.,quickly promoted to provide individual customer location change information positioning and navigation services.Based on the inertial sensor,the strapdown inertial navigation system has become a hotspot for each organization because of its simple structure,low cost and can solve the problem of positioning in extreme environment.Although the strapdown inertial navigation system is more accurate in the short term,it will accumulate error over time.However,the location fingerprinting technology based on wireless local area network does not generate the cumulative error,but because of the complexity and variability of the indoor environment,the positioning accuracy is not high to use the location fingerprint method based on WLAN alone.Therefore,how to combine inertial sensor-based navigation technology and WLAN-based location fingerprint positioning technology to play their own advantages,become the difficulty of the current study.In this paper,an improved maximum likelihood estimation fusion algorithm is proposed for the shortcomings of the existing maximum likelihood estimation fusion algorithm.Firstly,a zero-speed detection algorithm based on Kalman filter is proposed to solve the problem of cumulative error of strapdown inertial navigation.It lays the foundation for the research of fusion localization algorithm.Then,based on the inertial sensor PDR algorithm,it is necessary to set the initial coordinate and direction of the user or the problem of user intervention.In this paper,an improved initial state method is proposed based on the existing initial position determination method.Finally,a new maximum likelihood estimation fusion algorithm is proposed,which analyzes the defects from the characteristics of the maximum likelihood estimation fusion algorithm and the characteristics of PDR system and WiFi system location,and proposes an improved fusion based on maximum likelihood estimation algorithm,compared to the improved fusion algorithm,not only improve the positioning accuracy,the stability of the system has also been improved.The experimental results show that the proposed fusion positioning system can not only weaken the phenomenon of WiFi instability,but also solve the cumulative error of the PDR system based on inertial navigation and fix the positioning error caused by the initial position deviation.In the long time test experiment and in different environments still maintain a high degree of accuracy.