A new integer ambiguity estimation algorithm for single frequency precise point positioning

Abstract

In this paper, a new integer ambiguity resolution method for single-frequency PPP is proposed. In contrast to RTK method, a slightly perturbed integer value may incur large positioning error. To use carrier phase observations for positioning in PPP, it is necessary that the integer cycle is accurately fixed in order to have a highly precise location estimation. For these reasons, we proposed a new method to estimate the integer ambiguity. Firstly, SVD method is used to smooth the raw data, in which the code phase and carrier phase observations are both used. In this smooth process, some noise in the observations can be eliminated so that range of ambiguity search space is greatly reduced. We then estimate the floating point solution by using WLS method. Furthermore, in order to have more accurate integers, we proposed a modified LMS algorithm using a new cost function different from the existing method. From the results of the experiment, one can see that both sub-meters precision and accuracy are achievable using the proposed method.