BDS/WLAN Fusion Location Algorithm In Urban Complex Environment

With the demand for intelligent location-based services,spatial information has gradually become an indispensable and important element in our life.However,in the hot spot of human life such as urban environment,the BDS signal is often blocked and interfered,which leads to the problem that the positioning accuracy is reduced or even unable to be located.With the development of wireless positioning technology and the widespread popularity of WLAN access points in urban environment,WLAN positioning has gradually become a more mainstream indoor and urban positioning method.However,the fluctuation of received signal strength seriously affects the positioning accuracy of WLAN,which makes it impossible for WLAN to achieve positioning in complex urban environment alone.In response to this problem,this paper is focused on the location service needs of people in complex urban environments.Based on BDS positioning and WLAN location fingerprint location theory,aiming at improving the coverage and positioning accuracy of the positioning system,research work is carried out around the application of federated Kalman filtering and particle filtering in information fusion and positioning navigation.And in the current BDS/WLAN combination positioning,the corresponding solutions are proposed.The main research contents of this paper are as follows:(1)BDS/WLAN information fusion algorithm based on Federated Kalman.In the fusion of location information between BDS and WLAN,because WLAN cannot provide speed information,it cannot accurately construct the state equation needed in Kalman filtering algorithm.It will affect the final positioning results.To solve this problem,this paper designs a BDS/WLAN information fusion framework based on Federal Kalman according to its decentralized and centralized processing idea.In the process of model building,the following work has been accomplished.1)By analyzing the user's motion characteristics,the positioning error of pedestrians in three directions is taken as the state vector of the system,and the construction of the system state equation is further completed.2)The pseudo-range equation and the Doppler shift equation are combined with the single-difference principle to complete the construction of the measurement equation of the BDS subsystem.3)By analyzing the fingerprint localization algorithm of WLAN,it is proposed to establish the distance equation by using the distance obtained by online positioning matching combined with the access point and the position coordinates of the point to be measured.Thereby completing the construction of the measurement equation of the WLAN subsystem.In the process of building the model,the following work has been accomplished.(2)BDS/WLAN combined positioning algorithm based on improved particle filter.Particle filtering is a sequential Monte Carlo method to implement the recursive Bayesian estimation problem.It has better filtering effect in strong nonlinear and non-Gaussian systems.However,particle degradation is common in particle filter algorithms,which affects the positioning accuracy of BDS/WLAN combined systems.Aiming at this problem,this paper proposes a particle filter algorithm based on federated Kalman importance sampling,and completes the model construction.1)In this paper,the importance sampling function of particle filter is constructed by using the state information and covariance matrix obtained by the Federal Kalman Filter,so as to solve the problem of particle degradation.2)It is proposed to use the sampled particle position error obtained by the Federal Kalman as the weight to complete the particle update.Finally,experiments show that the proposed algorithm effectively implements the information fusion of BDS/WLAN,and improves the positioning accuracy of the integrated system.Therefore,the algorithm can be applied to location-based services in complex urban environments.