Comparison between CNMC and hatch filter & its precision analysis for BDS precise relative positioning

2015 
Phase smoothing Pseudorange can effectively suppress the multipath effect on pseudorange, and there is no ambiguity problem, which has been widely studied and applied in the field of GNSS data processing. This paper made a systematic analysis and comparison of mathematical models for Hatch filter and CNMC phase smoothing pseudorange method. The ionsphere free combination of single frequency pseudorange through CNMC algorithm is proved equivalence with dual frequency hatch filter. On the assumption that there exists only random errors and the various measurements are independent, the accuracy expression of the CNMC method for smoothing pseudorange is deduced. Dual band hatch filter in the initialization period observation noise is bigger than CNMC smoothed pseudorange, but after a bit more than ten minutes, observation noise gradually gets better than CNMC smoothed pseudorange. BDS navigation system is qualified with serious and low-frequency characteristic for multipath errors through the comparison of OMC of double difference observation for real BDS short baseline, the conclusion that the phase smoothing pseudorange algorithm can greatly weaken the random error of original pseudorange, however, can't improve the system error is drawn. The system error of pseudorange smoothed by CNMC is at the save level with the original pseudorange, but the system error of pseudorange smoothed by double frequency hatch filter is increased. At last, relative positioning is carried through using the different kinds of phase smoothed pseudorange of that short baseline. The positioning results show that the positioning precision is improved from 0.797 m to 0.541 m using pseudorange smoothed by CNMC and the positioning precision is decreased to 1.160 m using pseudorange smoothed by hatch filter.
    • Correction
    • Source
    • Cite
    • Save
    • Machine Reading By IdeaReader
    0
    References
    0
    Citations
    NaN
    KQI
    []