An Improved Two-Step Weighted Least Squares Aided UWB High-Precision Indoor Positioning System for Moving Targets

In recent years, high-precision indoor positioning (HPIP) technologies are of considerable interest due to the wide range of applications within the Internet of Things area. However, the performance of most positioning systems highly depends on the estimate of initial position, i.e. when the initial position estimate is inaccurate, the algorithms may not converge, which significantly limits the positioning accuracy and system efficiency. This work proposes an improved two-step weighted least squares (ITSWLS) aided ultra-wideband (UWB) indoor positioning system, where in the first stage, the proposed ITSWLS forms a set of pseudo-linear equations by introducing the nuisance parameters, i.e. a zero-mean Gaussian distributed noise, and then it applies a weighted least squares (WLS) minimization to the problem for obtaining an initial solution. In the second stage, the relationship among the nuisance parameters and the position-velocity of the moving target are both utilized for refining the initial solution of the first stage for improving the overall system’s performance. Additionally, the Kalman filtering and the time difference of arrival and frequency difference of arrival (TDoA-FDoA) positioning framework are utilized for the sake of further improving the positioning accuracy performance. Finally, the proposed HPIP system is implemented into a practical positioning hardware system. Both the simulation results and the practical hardware tests are presented to demonstrate the validity and effectiveness of the proposed HPIP system.