An improved algorithm based on combination observations for real time cycle slip processing in triple frequency BDS measurements

摘要

This research presents a new algorithm to resolve real-time cycle slip for undifferenced triple frequency observations of BeiDou navigation satellite system (BDS). The method of cycle slip (CS) detection is based on three linearly independent combinations: GIF (geometry free and ionosphere free) combination, GF (geometry free) phase combination and code-phase combination. First order differenced ionospheric delay is predicted by a back moving window filter. The difference between predicted and calculate first order differenced ionospheric delay is adopted to detect the insensitive cycle slips and to judge whether the detected cycle slips are repaired correctly. With respect to cycle slip determination, modified LAMBDA (MLAMBDA) method is adopted. The proposed method has been verified by real BDS observations adding various artificial cycle slips at every ten epochs. The results show that the proposed algorithm is reliable to detect and repair small and large cycle slips even under low sampling interval (30 s). The fix rate of CS can reach 100% for all the selected satellites. For the most insensitive cycle slip group (1, 1, 1), the fixing rate can reach 99.82% without considering the first order differenced ionospheric delay condition. The fixing rate can also reach 100% when ionospheric delay condition is considered. In terms of the comparative method, the fixing rate is about 91% in according to the IGSO satellites. These results demonstrate that the proposed algorithm can detect and repair all added cycle slips even under low sampling rate at 30 s. (C) 2018 COSPAR. Published by Elsevier Ltd. All rights reserved.