Article Details

Abstract

To address the problems of the traditional Melbourne–Wübbena (MW) method's susceptibility to pseudorange noise interference, missed detection of minor cycle slips, and divergence after threshold failure, this paper proposes a MW cycle slip detection method based on Long Short-Term Memory (LSTM) neural network denoising. The method leverages LSTM networks to learn temporal characteristics and noise distribution patterns in cycle-slip-free observation data, enabling noise suppression for MW combined values and employing a residual fixed threshold method for cycle slip discrimination. Using BeiDou C40 satellite data, artificial cycle slip injection simulations are conducted. Results show that, compared to the traditional MW method, the noise fluctuations in the denoised MW detection metrics are significantly reduced, with the detection threshold dropping to 0.215 cycles and the cycle slip missed detection rate decreasing from 50% to 0%, effectively improving the accuracy and reliability of cycle slip detection.

Keywords

BeiDou; cycle slip detection; Melbourne–Wübbena combination; LSTM

References

1. [1] FAROOQ S Z，YANG D K，ADA E N J. A Cycle Slip Detection Framework for Reliable Single Frequency RTK Positioning [J/OL].Sensors，2020，20（1）：304[2026-04-18].https://doi.org/10.3390/s20010304.
2. [2] ZHAO X Y，NIU Z，LI G，et al. A New Cycle Slip Detection and Repair Method Using a Single Receiver’s Single Station B1 and L1 Frequencies in Ground-Based Positioning Systems [J/OL].Sensors，2020，20（2）：346[2026-04-18].https://doi.org/10.3390/s20020346.
3. [3] Liu Ning, Liu Shihuan, Wu Xiaoli, etc Universal GNSS Dual Frequency Data Cycle Jump Detection and Repair Method [J]. Geodesy and Geodynamics, 2024, 44 (10): 1003-1009
4. [4] Xia Lei, Lai Zulong, Mei Changsong, etc TurboEdit epoch difference optimization algorithm for real-time cycle slip detection and repair [J]. Journal of Wuhan University (Information Science Edition), 2021, 46 (6): 920-927
5. [5] Jia Qianzi, Gao Ling, Liu Mutian Overview of GNSS cycle slip detection [J]. Radio Engineering, 2025, 55 (8): 1658-1666
6. [6] Huang Yuanyuan, Zhao Dangli, Zhao Shujie, etc Research on the joint detection and repair of BDS-3 carrier phase cycle slip using MW and STPIR combination [J]. Journal of Time and Frequency, 2023, 46 (4): 333-340
7. [7] JIAO Y X，LI K Z，YUE Z，et al. CEEMDAN Combined Wavelet Denoising Improved the MW Cycle Slip Detection Algorithm [J].Scientific Reports，2024，14（1）：23487-23487.
8. [8] NING Y P，CUI J Y，ZHAO W S，et al. A Novel Cycle Slips Detection and Repair Method with AR Model of BDS-3 Dual-Frequency Signal in Severe Multipath Environments [J/OL].Applied Sciences，2022，13（1）：27[2026-04-18].https://doi.org/10.3390/app13010027.
9. [9] Jing Xin, Li Jianwen, Li Linyang, etc BDS cycle slip detection algorithm with total variation MW combination denoising [J]. Journal of Navigation and Positioning, 2023, 11 (2): 176-185
10. [10] Liang Lingfeng Research on GNSS cycle slip detection and repair using single frequency double difference LM-BP neural network [J]. Modern Information Technology, 2022, 6 (15): 56-58
11. [11] Dong Ming, Zhang Yinping Application of Wavelet LSTM Neural Network Model in Time Series Prediction of Settlement Monitoring [J]. Surveying and Spatial Geographic Information, 2025, 48 (11): 157-159

Funding

Major Science and Technology Projects of Anhui Province in 2021 (202103a05020026); 2021 Anhui Province Key Research and Development Plan (202104a07020014)

About the authors

Sheng Minghong (1973-), female, Han ethnicity, engineer, bachelor's degree, research direction: Geodesy and Surveying Engineering.