Testing of domestic software-mathematical complex «OCTAVA» for centimeter accuracy GNSS positioning – network processing of static and kinematic L1-observations obtained at baselines up to ~100 km

Kharkov National University of Radio Electronics (KhNURE)

Scientific-Educational Center (SEC) of sub-department «The Fundamentals of Radio Engineering»

Scientific-Research Laboratory (SRL)

«Satellite Network Technologies of High Precision Positioning» («SNTPP»)


Tel.: +38(057) 700-22-84, E-mail: This email address is being protected from spambots. You need JavaScript enabled to view it. www.KharkovGNSSgroup.net


Alexey A. Zhalilo, Alexey A. Zhelanov, Igor V. Ditskiy, Evgeniy A. Bessonov


(Summary of the scientific-technical report, Kharkov, April-May 2013)


Scientific-technical report includes: 115 pages, 45 figures, 25 tables, 15 references, 5 appendices ( 60 pages).

In the report are presented the results of testing the developments of the SRL «SNTPP» – methods, algorithms and software for post processing of dual frequency and single frequency network GNSS (GPS) observations for solving the tasks of centimeter accuracy positioning in static and kinematic surveying. The purpose of testing – fine-tuning of new effective methods of processing the observations, demonstration of performance capabilities and comparison the features of the developed prototype of the software complex «OCTAVA» with the existing foreign products-analogues. It is shown that the developed domestic technology of high precision positioning is competitive, comparable by its accuracy characteristics with the best foreign analogues, and in part of single frequency positioning – excels the foreign analogues in class of software products for post processing.

In the course of experiments and comparative analysis the following main results were obtained.

  1. It is shown that the characteristics of the developed technology of dual frequency differential positioning are adequate to the characteristics of the tested foreign analogue.
  2. During single frequency positioning in a static mode of survey in the zone of the network with baselines up to 200 km at distances from the nearest station ~100 km the values of positioning errors made up less 1 cm (RMS) in horizon plane and ~1-2 cm (RMS) in local vertical (the required interval of observation – 30-40 minutes).
  3. During single frequency positioning in a kinematic mode in the zone of the network with baselines up to 200 km at distances from the nearest station ~100 km the single frequency positioning errors of the aircraft (helicopter) carrying out the aerosurveying made up ~3-4 cm (RMS) in horizon plane and ~5-6 cm (RMS) in local vertical. For reliable carrier-phase ambiguity resolution there are required 50-60 minutes of continuous observations without the static initialization.
    In the zone of the network with baselines up to ~100-120 km the single frequency positioning errors made up ~1-2 cm (RMS) in plane and ~3-4 cm (RMS) in vertical.
  4. The comparison of the accuracy of kinematic positioning using the developed post processing single frequency technology and the accuracy of the well-known network dual frequency RTK technologies (VRS, MAC, FKP) has shown that for networks with baselines ~100 km these technologies are comparable in accuracy. While using the networks with baselines ~150-200 km the domestic single frequency technology provides the better accuracy results.

  5. Keywords: global navigation satellite systems (GNSS), Global Positioning System (GPS), base (reference) GNSS stations, code and carrier phase GNSS observations/measurements, carrier phase ambiguity resolution, precise positioning, ionospheric delay, zenith tropospheric delay, interpolation, postprocessing, software development kit (SDK).


    Example

    Network Map GNSS stations

    Figure 1 – Helicopter track and reference station network

    altitude_change

    Figure 2 – Helicopter height variation during the flight

    Fig. 3 shows the dependence of static positioning errors as function from the interval of observations at baselines ~70 km (figure on the left) and ~160 km (figure on the right).


    coordinate_error_1 coordinate_error_2

    Figure 3 – Dependence of static L1–solution errors as function from the interval of observations

    coordinate_error_3

    Figure 4 – Discrepancies (differences) of coordinate estimations of the network single frequency and dual frequency solutions in a mode of kinematic survey (helicopter) - the reference station network with baselines up to ~100 km

    Table 1 – Statistical characteristics of current deviations of coordinates of the network single frequency and dual frequency solutions in a mode of kinematic survey (helicopter) - the reference station network with baselines up to ~100 km

    Coordinates Er (P=68%), м Er (P=95%), м Er (P=99.7%), м
    X 0.026 0.043 0.075
    Y 0.013 0.030 0.039
    Z 0.022 0.044 0.066
    B 0.022 0.039 0.053
    L 0.014 0.029 0.041
    H 0.030 0.058 0.088

    coordinate_error_4

    Figure 5 – Discrepancies (differences) of coordinate estimations of the network single frequency and dual frequency solutions in a mode of kinematic survey (helicopter) - the reference station network with baselines up to ~200 km

    Table 2 – Statistical characteristics of current deviations of coordinates of the network single frequency and dual frequency solutions in a mode of kinematic survey (helicopter) - the reference station network with baselines up to ~200 km


    Coordinates Er (P=68%), м Er (P=95%), м Er (P=99.7%), м
    X 0.046 0.110 0.136
    Y 0.026 0.051 0.062
    Z 0.042 0.096 0.122
    B 0.035 0.072 0.094
    L 0.029 0.057 0.066
    H 0.055 0.135 0.170

    For the purpose of comparing the results of testing of the developed technologies with typical statistical characteristics of RTK-positioning errors there have been used the results obtained in the paper [*]. From this paper there have been taken the results of accuracy estimations (see table 3) for three widely-used network dual frequency RTK-technologies: VRS, MAC и FKP. The tests were carried out in Italy for three configurations of networks with different baselines: 50 km, 100 km and 150 km.


    Table 3 – Statistical characteristics of positioning errors for different dual frequency network RTK technologies (see [*])

    Networks (typical interstation distances) 50 км 100 км 150 км 50 км 100 км 150 км 50 км 100 км 150 км
    Er (P=68%), м Er (P=95%), м Er (P=99.7%), м
    NRTK VRS
    В, L, см 1.0 2.0 5.0 3.0 7.5 17.5 3.5 20.0 30.0
    H, см 2.0 4.0 11.0 4.5 11.0 23.5 6.5 22.0 38.0
    NRTK MAC
    В, L, см 3.0 1.0 2.5 7.0 7.0 15.0 34.0 16.0 24.0
    H, см 6.5 2.0 4.0 15.5 14.0 24.0 52.0 20.0 40.0
    NRTK FKP
    В, L, см 5.0 15.0 18.0 15.0 27.0 25.0 32.0 32.0 30.0
    H, см 7.5 10.0 15.5 15.5 28.0 34.0 36.0 50.0 55.0

    [*] - Paolo Dabove, Mattia De Agostino Network RTK and Reference Station Configuration // InsideGNSS, USA, November/December 2011, v.6, №6. – pp. 24-29.


    Download the abstract [Ru] Download the abstract [En] Download the presentation [Ru]