Assessment of strategies for detection and correction of cycle slips on GPS L1 observables

Abstract

To prevent large errors in the GPS positioning, cycle slips should be detected and corrected. Such procedure is not trivial, mainly for single frequency receivers, but normally it is not noticed by the users. Thus, it will be discussed some practical and more used methods for cycle slips detection and correction using just GPS single-frequency observations. In the detection, the triple (TD) and tetra differences were used. In relation to the correction, in general, each slip is corrected in the pre-processing. Otherwise, other strategies should be adopted during the processing. In this paper, the option was to the second option, and two strategies were tested. In one of them, the elements of the covariance matrix of the involved ambiguities are modified and new ambiguity estimation starts. In the one, a new ambiguity is introduced as additional unknown when a cycle slip is detected. These possibilities are discussed and compared in this paper, as well as the aspects related to the practicity, implementation and viability of each one. Some experiments were carried out using simulated data with cycle slips in different satellites and epochs of the data. This allowed assessing and comparing the results of different occurrence of cycle slip and correction in several conditions.