Multi-frequency BeiDou cycle slip and data gap repair with geometry-based model

In real GNSS high-precision applications with carrier phase observations, the cycle slips and short-time interruptions often occur, leading to a long time re-initialization. The traditional methods often process cycle slips of satellite-by-satellite based on the geometry-free combinations of multiple frequencies. The geometry-free model ignores the mutual strong links between different satellite observations, which degrades the efficiency of cycle slip processing. With gradual availability of multi-frequency GNSS, the correlations of observations between multi-frequency and between satellites should be taken full advantage of. In this contribution, we will dedicate to present a geometry-based and ionosphere-weighted approach for integer cycle slip estimation, along with handling of receiver coordinate parameters, receiver clock error and ionospheric delays, which is generally applicable to arbitrary number of frequencies of CDMA satellite systems. When it is unable to effectively fix all of the cycle slips simultaneously, we further propose to partially fix the cycle slips that can be reliably fixed. Extensive experiments are carried out using undifferenced dual/triple-frequency BDS data to validate the performance of our proposed method. The result shows that by using the predicted ionospheric variations, the data gaps up to 30-45 s can be effectively connected, and the cycle slips can be correctly fixed even in case of the poor situations where continuous cycle slips occur on all satellites.