A Data Fusion Scheme for Modified EKF Banks Positioning Algorithm in Mixed LOS/NLOS Conditions

A data fusion scheme of low computational complexity and high estimation accuracy is very important for mobile location. In this paper, a novel data fusion scheme that incorporates both pre-processing before data fusion and post-processing after fusion is proposed to modified EKF banks positioning algorithm in mixed LOS/NLOS conditions via the navigation information. The pre-processing module can intelligently select the local location estimates for data fusion. As some local estimates with large errors are deleted, the overall computational complexity is reduced while the location accuracy is improved. The data post-processing is implemented by a position outlier detection and correlation which is to remove and correct the fused position outliers so as to further improve the fusion performance. Simulation results show that the proposed data fusion scheme outperforms the traditional method in terms of the estimation accuracy and computational complexity, when modified EKF banks positioning algorithm is utilized.