In case of indoor or outdoor positioning, the Gauss-Newton (GN) method has a divergence problem while solving user position iteratively, especially in poor dilution of precision (DOP) condition. To solve that, quadratic correlated least square (QCLS) algorithm is introduced. In this paper, the ranging measurement is assumed to be time of arrival (TOA). Squared measurement contains second-order term which cannot be treated as linear term. By setting other variable or differencing between squared measurements, user position could be obtained. However, the aforementioned solution loses some of information contained in the ranging measurements and should be corrected in order to acquire precise and robust solution. This paper suggests QCLS algorithms using TOA measurement to correct previous coarse estimation. The performance of QCLS algorithm is compared with other algorithms by using experimental data.

Keywords: two-way ranging, modified QCLS, indoor positioning, CSS-UWB