NLOS error mitigation with information fusion algorithm for UWB ranging systems

This article puts forward a scalar weighting information fusion （IF） smoother with modified biased Kalman filter （BKF） and maximum likelihood estimation （MLE） to mitigate the ranging errors in ultra wide band （UWB） systems. The information fusion algorithm uses both the time of arrival （TOA） and received signal strength （RSS） measurement data to improve the ranging accuracy. At first, the ranging protocol of IEEE 802.15.4a acts as a multi-sensor system with multi-scale sampling. Then the scalar-based IF smoother accurately estimates the range measurement in the line of sight （LOS） and non-line of sight （NLOS） condition of UWB sensor network, during which the effectiveness of the IF in mitigating errors is especially focused during the LOS/NLOS transitions. Simulation results show that the proposed hybrid TOA-RSS fusion approach indicates a performance improvement compared with the usual TOA-only and other IF method, and the estimated ranging metrics can be used for achieving higher accuracy in location estimation and target tracking.

A novel TOA estimation method with effective NLOS error reduction

It is well known that non-line-of-sight （NLOS） error has been the major factor impeding the enhancement of accuracy for time of arrival （TOA） estimation and wireless positioning. This article proposes a novel method of TOA estimation effectively reducing the NLOS error by 60%, comparing with the traditional timing and synchronization method. By constructing the orthogonal training sequences, this method converts the traditional TOA estimation to the detection of the first arrival path （FAP） in the NLOS multipath environment, and then estimates the TOA by the round-trip transmission （RTT） technology. Both theoretical analysis and numerical simulations prove that the method proposed in this article achieves better performance than the traditional methods.