Cycle slip detection and repair based on the unmodeled-error-constrained geometry-free combining geometry-based models for a single-frequency receiver

Abstract

Accurate cycle slip detection and repair is the prerequisite of high-precision and high-reliability Global Navigation Satellite System (GNSS) positioning and navigation. The cycle slips are widely existent in a challenging environment or a low-cost receiver, where the unmodeled errors are usually significant. Therefore, it is highly urgent to enhance the performance of cycle slip detection and repair for a single-frequency receiver under complex conditions. In this paper, a new cycle slip processing method based on the unmodeled-error-constrained geometry-free (GF) combining geometry-based (GB) models is proposed. Specifically, first, the short-term GF and GB unmodeled errors are predicted based on their temporal behaviours. Second, the predicted GF and GB unmodeled errors are applied to correct the potentially biased GF and GB models. Finally, the cycle slips are detected and repaired by combing the unmodeled-error-constrained GF and GB models. By using the real dataset with multipath effects and ionospheric delays, the proposed short-term prediction of unmodeled errors is effective both in GF and GB models, where the centimetre-level accuracy can be obtained, respectively. Based on this, three different typical situations, i.e., large, small, and mixed cycle slips, are all analysed. The results show that compared with the traditional methods, the proposed method considering the unmodeled errors is the most effective. Precisely, it can process the cycle slips with a detection percentage of 100% and a mean repair percentage of 93% even under challenging environments.