A Real-Time Cycle Slip Detection and Repair Method Based on Ionospheric Delay Prediction for Undifferenced Triple-Frequency BDS Signals

Abstract

The detection and repair of cycle slips are key steps in high-accuracy GNSS (Global Navigation Satellite System) data processing using carrier phase observations. BDS (BeiDou Navigation Satellite System) triple-frequency observations provide better combinations for cycle slip detections and repairs compared to dual-frequency observations. Although a number of algorithms have been developed and may correctly detect cycle slips most of the time, the reliability of empirical thresholds methods cannot be guaranteed. In this study, an adaptive threshold is proposed for three sets of triple-frequency Geometry-Free (GF) pseudorange minus phase combinations to improve the cycle slip detection performance and reduce the false alarm rate of the cycle slip detection by combining the predicted epoch-differenced ionospheric delays under active ionospheric conditions. Moreover, in the cycle slip repair, the integral combined cycle slips are determined by directly rounding the estimated float-combined cycle slips, which will lead to a repair error if the between-epoch ionospheric variation is large. In this study, a new rounding method considering the predicted epoch-differenced ionospheric delays is proposed, and it is proven that the new method has a higher success rate for estimating the integer value of a cycle slip than the traditional method. The performance of the newly proposed method is validated by using static BDS triple-frequency observations that contain simulated cycle slips. BDS triple-frequency observations were collected at 30-s sampling intervals under active ionospheric conditions. The results show that this method can successfully detect and repair all slips of more than one cycle. In addition, dynamic BDS data collected with a vehicle-based receiver at a 1-s sampling intervals are processed, and the results show that the proposed method is also effective in the detection and repair of cycle slips in dynamic data.