Determination of optimal parameters for GPS/GNSS cycle slip detection algorithm

Abstract Accurate detection of cycle slips is vital for geodetic GNSS-based positioning with carrier phase data, and for quality control of GNSS data processing for various applications. Typically, cycle slip detection algorithms use certain parameters, such as a scale factor of the root-mean-square (RMS) of residuals of linear combinations and an ionospheric threshold value, to flag potential cycle slips. The performance of an algorithm will depend on the values assigned for such parameters. It is important that the parameter values are optimized such that the algorithm minimizes the rates of both missed slips and false detections. We evaluate two parameters (a scale factor, k , of Hatch-Melbourne-Wübbena and ionosphere-free residual RMS and an ionospheric residual threshold, h ) to determine their optimal value for cycle slip detection. We show that in the worst-case scenario involving only small slips of magnitudes ranging from 1 to 3 cycles, there is an optimal range of values which resulted in 93 to 97% success rate without any false detections. For the data quality considered in this study, we recommend an RMS scale factor of k = 6 and an ionospheric residual threshold of h = 0.3 as the optimal values for the cycle slip detection with at least 97% success rate. In the absence of formal confidence limits, we note that combinations of values in the range of 6 ≤ k ≤ 8 and 0.3 ≤ h ≤ 1.0 resulted in at least 90% success rate with zero false detections, which gives some indication of the sensitivity of the algorithm’s performance to the specific choice of the parameter values.