Single-station single-frequency GNSS cycle slip estimation with receiver clock error increment and position increment constraints

Abstract

Cycle slip detection is essential for achieving centimeter-level positioning using Global Navigation Satellite Systems (GNSS). However, when dealing with single-frequency data, it is impossible to utilize observations from multiple frequencies to construct effective linear combinations for detecting cycle slips. Moreover, the data quality of low-cost receivers is relatively poor, making the process more challenging. In this study, a novel technique is presented for detecting single-frequency cycle slips in a single receiver. The approach includes additional constraints for position and clock error increments. By leveraging the random walk characteristics, the clock error increment is predicted, and the cycle slip detection term is then formulated using the position increment constraint of the odometer. Both static and dynamic experiments demonstrate that the detection term’s three times standard deviation is less than 0.2 cycles. Furthermore, the method can achieve clock error increment accuracy of 6.9 mm and 3.2 mm in situations where traditional TDCP technology fails under 2 and 3 visible satellites condition, respectively. This represents a 22.47% and 64.04% improvement over the accuracy of direct prediction from the previous epoch. It avoids long-term prediction of clock error increment until divergence in complex environments and maintains the continuity of cycle slip detection. In addition, we explore the clock error increment characteristics of 10 types of receivers in 6 datasets, providing a new consideration index for the popularization of low-cost GNSS receivers from the perspective of receiver type selection.