GNSS signal processing with rover ambiguity setting

The MW bias processing and Kalman filter method addresses ambiguity resolution issues in GNSS systems, improving precision and real-time accuracy by estimating satellite biases and ambiguities, particularly in GLONASS, through ionosphere-free and widelane ambiguity resolution.

DE112010003672B4Active Publication Date: 2024-11-28TRIMBLE INC
View PDF 15 Cites 0 Cited by

Patent Information

Application Number
DE112010003672
Authority / Receiving Office
DE · DE
Patent Type
Patents
Current Assignee / Owner
Priority Date
2009-09-19
Filing Date
2010-09-19
Publication Date
2024-11-28
Estimated Expiration
2030-09-19

AI Technical Summary

Technical Problem

Existing GNSS signal processing methods struggle with ambiguity resolution, particularly in systems like GLONASS, where hardware-induced clock errors and frequency-dependent distortions complicate the estimation of carrier ambiguities, affecting precision and real-time positioning accuracy.

Method used

A method involving the use of Melbourne-Wübbena (MW) bias processing and Kalman filters to estimate satellite biases and ambiguities, incorporating ionosphere-free and widelane ambiguities, and applying linear combinations to resolve integer ambiguities in GNSS signal data, enhancing precision and real-time processing capabilities.

Benefits of technology

This approach improves the accuracy and efficiency of GNSS signal processing by effectively resolving ambiguities, enabling precise positioning even in challenging environments with hardware-induced errors and frequency distortions, thus enhancing the reliability of real-time navigation systems.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure 00000000_0000_ABST
    Figure 00000000_0000_ABST
Patent Text Reader

Abstract

A method for processing a set of GNSS signal data derived from signals of a set of satellites with carriers observed at a rover antenna, wherein the data include a carrier observation and a code observation of each carrier from each satellite, comprising: Received for each satellite time signal corrections, comprising at least two of: (i) a code-matched satellite time signal, (ii) a phase-matched satellite time signal, and (iii) a satellite time signal bias representing a difference between a code-matched satellite time signal and a phase-matched satellite time signal, Running a first filter that uses at least the GNSS signal data and the satellite time signal corrections to estimate values ​​for parameters, comprising at least one carrier ambiguity for each satellite and a covariant matrix of carrier ambiguities, wherein the code-matched satellite time signal is used to model all GNSS observations, and a floating ambiguity is adjusted to the level of the phase-matched time signal by applying the difference between the code-matched satellite time signal and the phase-matched satellite time signal. Determining each carrier ambiguity of an integer nature, comprising one of: an integer value, and a combination of integer candidates, inserting the carrier ambiguities of an integer nature as pseudo-observations into a second filter, and applying the second filter to the GNSS signal data and the satellite time signal corrections to obtain estimated values ​​for parameters, comprising at least the receiver position.
Need to check novelty before this filing date? Find Prior Art

Description

BACKGROUND

[0001] The invention relates to GNSS signal processing and in particular to GNSS signal processing with precise satellite data.

[0002] WO 2008 / 051201 A2 describes methods for sliding solutions and ambiguity resolution for multiple global navigation satellite systems (GNSS), one of which may be an FDMA-based GNSS such as GLONASS. This includes modeling the hardware-induced differential clock error between different GNSS systems, modeling the frequency-dependent distortions present in frequency-division multiplexed (FDMA) GNSS, and a method for resolving ambiguities called scoreboard partial fixing (SPF). The methods are independent of the number of carrier frequencies tracked for each satellite navigation system. BRIEF SUMMARY

[0003] A method and apparatus are described for processing a set of GNSS signal data derived from signals of a set of satellites with carriers, observed at a rover antenna, wherein the data include a carrier observation and a code observation of each carrier from each satellite, comprising: Obtain for each satellite time signal corrections, comprising at least two of: (i) a code-matched satellite time signal, (ii) a phase-matched satellite time signal, and (iii) a satellite time signal bias representing a difference between a code-matched satellite time signal and a phase-matched satellite time signal; run a first filter, using at least the GNSS signal data and the satellite time signal corrections, to estimate values ​​for parameters, comprising at least one carrier ambiguity for each satellite and a covariant matrix of carrier ambiguities; determine each carrier ambiguity of an integer nature, comprising one of: an integer value and a combination of integer candidates.Inserting the carrier ambiguities of an integer nature as pseudo-observations into a second filter and applying the second filter to the GNSS signal data and the satellite time signal corrections to obtain estimated values ​​for parameters, including at least the receiver position.

[0004] In some embodiments, the carrier ambiguities are of an integer nature, and inter-satellite ambiguities are single-differentiated ambiguities. Some embodiments further include obtaining a set of MW corrections, running a third filter using the GNSS signal data and at least the MW corrections to obtain at least one set of WL ambiguities, and using the set of WL ambiguities to obtain the carrier ambiguity of an integer nature. In some embodiments, the WL ambiguities include at least one of: float values, integer values, and float values ​​based on integer candidates. In some embodiments, the covariant matrix of ambiguities is scaled to reflect the change due to the use of the WL ambiguities. In some embodiments, the WL ambiguities are inter-satellite ambiguities.

[0005] In some embodiments, the integer ambiguities include at least one of: L1-L2 ionospheric-free ambiguities, L2-L5 ionospheric-free ambiguities, and carrier ambiguities of a linear combination of two or more GNSS frequencies. In some embodiments, the ionospheric delay information is used to feed to one or more of the filters, and the integer ambiguity includes at least one of: the carrier ambiguity of L1 frequency, the carrier ambiguity of L2 frequency, the carrier ambiguity of L5 frequency, and the carrier ambiguity of any GNSS frequency. In some embodiments, the second filter includes one of: a new filter, a copy of the first filter, and the first filter.In some embodiments, the code-matched satellite time signal is used to model all GNSS observations, and the floating ambiguity is formed at the level of the phase-matched time signals by applying the difference between the code-matched satellite time signal and the phase-matched satellite time signal.

[0006] In some embodiments, the code-matched satellite time signal is used to model all GNSS code observations, and the phase-matched satellite time signal is used to model all GNSS carrier observations. In some embodiments, determining the carrier ambiguity of an integer nature from a floating ambiguity includes at least one of: rounding the floating ambiguity to the nearest integer, selecting best integer candidates from a set of integer candidates generated using integer least squares, and calculating floating values ​​using a set of integer candidates generated using integer least squares.In some embodiments, at least one of the first filter and the second filter further estimates at least one of a receiver-phase-matched time signal, receiver-code-matched time signal, tropospheric delay, receiver-time signal bias representing a difference between the code-matched receiver time signal and the phase-matched receiver time signal, and multipath states. In some embodiments, at least one of the first filter, the second filter, and the third filter is configured to update the estimated values ​​for each of a plurality of epochs of GNSS signal data.

[0007] Some embodiments provide a device for performing one or more of the described methods. Some embodiments provide a computer program comprising instructions, configured when executed on a computer processing unit, to perform one or more of the described methods. Some embodiments provide a physical, computer-readable medium embodying such a computer program. BRIEF DESCRIPTION OF THE DRAWINGS

[0008] These and other aspects and features of the present invention will be more quickly understood from the embodiments described below with reference to the drawings, showing the following: Fig. Figure 1 shows a high-level view of a system according to some embodiments of the invention; Fig. Figure 2 shows a high-level view of a system and system data according to some embodiments of the invention; Fig. Figure 3 shows a schematic diagram of a network processor architecture according to some embodiments of the invention; Fig. Figure 4 shows a schematic diagram of a data correction according to some embodiments of the invention; Fig. Figure 5 shows a schematic view of a linear combination of observations according to some embodiments of the invention; Fig. Figure 6 shows a schematic view of a generic Kalman filter process; Fig. Figure 7 shows a schematic diagram of a code-adapted time signal processor or clock processor according to some embodiments of the invention; Fig. 8, Fig. 9 and Fig. 10 have been deleted; Fig. Figure 11 shows a schematic diagram of a Melbourne-Wübbena bias process flow according to some embodiments of the invention; Fig. Figure 12 shows a schematic diagram of a Melbourne-Wübbena bias process flow according to some embodiments of the invention; Fig. Figure 13A shows filter states of an undifferentiated Melbourne-Wübbena bias processor according to some embodiments of the invention; Fig. Figure 13B shows filter states of an undifferentiated Melbourne-Wübbena bias processor according to some embodiments of the invention; Fig. Figure 14 shows a schematic diagram of a Melbourne-Wübbena bias processor according to some embodiments of the invention; Fig. Figure 15 shows a schematic diagram of a Melbourne-Wübbena bias processor according to some embodiments of the invention; Fig. Figure 16 shows a schematic diagram of a Melbourne-Wübbena bias processor according to some embodiments of the invention; Fig. Figure 17 shows a schematic diagram of a Melbourne-Wübbena bias processor according to some embodiments of the invention; Fig. Figure 18 shows a schematic diagram of a Melbourne-Wübbena bias processor according to some embodiments of the invention; Fig. Figure 19A shows an observation graph of GNSS stations and satellites; Fig. Figure 19B shows an abstract graph that represents stations and satellites as nodes and vertices respectively, and station-satellite observations as edges; Fig. Figure 19C shows a minimum spanning tree of the graph of Fig. 19B; Fig. 19D shows a minimum spanning tree with restricted edges; Fig. Figure 19E shows an undifferentiated observation graph of GNSS stations and satellites; Fig. Figure 19F shows a filter graph that corresponds to the observation graph of Fig. 19E corresponds to; Fig. Figure 19G shows a detailed observation graph of GNSS stations and satellites; Fig. Figure 19H shows a filter graph that corresponds to the observation graph of Fig. 19G corresponds to; Fig. Figure 19I shows a set of observation graphs that compares conditions or boundary conditions in undifferentiated and individually differentiated processing; Fig. Figure 20 shows a schematic diagram of a Melbourne-Wübbena bias processor according to some embodiments of the invention; Fig. Figure 21A shows a spanning tree of an undifferentiated observation graph; Fig. Figure 21B shows a minimum spanning tree on an undifferentiated observation graph; Fig. Figure 21C shows a spanning tree on a single-differentiated observation graph; Fig. Figure 21D shows a minimum spanning tree on a single-differentiated observation graph; Fig. Figure 22 shows a schematic diagram of a Melbourne-Wübbena bias processor according to some embodiments of the invention; Fig. Figure 23A shows a schematic diagram of a Melbourne-Wübbena bias processor according to some embodiments of the invention; Fig. Figure 23B shows a schematic diagram of a Melbourne-Wübbena bias processor according to some embodiments of the invention; Fig. Figure 24A shows a schematic diagram of a Melbourne-Wübbena bias processor according to some embodiments of the invention; Fig. Figure 24B shows a schematic diagram of a Melbourne-Wübbena filter process according to some embodiments of the invention; Fig. Figure 24C shows a schematic diagram of a Melbourne-Wübbena filter process according to some embodiments of the invention; Fig. Figure 24D shows a schematic diagram of a Melbourne-Wübbena filter process according to some embodiments of the invention; Fig. Figure 25A shows a schematic diagram of a Melbourne-Wübbena bias processor according to some embodiments of the invention; Fig. 25B represents the effect of a shifting of biases according to some embodiments of the invention; Fig. Figure 25C shows a schematic diagram of a Melbourne-Wübbena bias processor according to some embodiments of the invention; Fig. Figure 26A shows a schematic diagram of starting an orbit processor according to some embodiments of the invention; Fig. Figure 26B shows a schematic diagram of an orbit processor according to some embodiments of the invention; Fig. Figure 26C shows a schematic diagram of an orbit mapper of an orbit processor according to some embodiments of the invention; Fig. Figure 26D shows a schematic diagram of an orbit mapper of an orbit processor according to some embodiments of the invention; Fig. Figure 27 shows a timing diagram or clock diagram of a code-adapted time signal processing according to some embodiments of the invention; Fig. Figure 28A shows a schematic diagram of a code-adapted high-rate satellite time signal processor according to some embodiments of the invention; Fig. Figure 28B shows a schematic diagram of a code-adapted high-rate satellite time signal processor according to some embodiments of the invention; Fig. Figure 28C shows a schematic diagram of a code-adapted high-rate satellite time signal processor according to some embodiments of the invention; Fig. Figure 29 shows a schematic diagram of a phase-matched high-rate satellite time signal processor according to some embodiments of the invention; Fig. Figure 30A shows a schematic diagram of a phase-matched high-rate satellite time signal processor according to some embodiments of the invention; Fig. Figure 30B shows a schematic diagram of a phase-matched high-rate satellite time signal processor according to some embodiments of the invention; Fig. Figure 30C shows a schematic diagram of a phase-matched high-rate satellite time signal processor according to some embodiments of the invention; Fig. Figure 31 shows a schematic diagram of a phase-matched high-rate satellite time signal processor according to some embodiments of the invention; Fig. Figure 32 shows a schematic diagram of a phase-matched high-rate satellite time signal processor according to some embodiments of the invention; Fig. Figure 33 shows a schematic diagram of a phase-matched high-rate satellite time signal processor according to some embodiments of the invention; Fig. Figure 34 shows a schematic diagram of a phase-matched high-rate satellite time signal processor according to some embodiments of the invention; Fig. 35 is deleted Fig. Figure 36 shows a schematic diagram of a network processor computer system according to some embodiments of the invention; Fig. Figure 37 shows a simplified schematic diagram of an integrated GNSS receiver system according to some embodiments of the invention; Fig. Figure 38 shows a schematic diagram of a GNSS rover process with synthesized base station data according to some embodiments of the invention; Fig. Figure 39 shows an observation time prediction or observation time signal prediction according to some embodiments of the invention; Fig. Figure 40 shows a schematic diagram of a process for generating synthesized base station data according to some embodiments of the invention; Fig. 41 is deleted Fig. Figure 42 shows a schematic diagram of an alternative GNSS rover process using synthesized base station data according to some embodiments of the invention; Fig. Figure 43 shows a simplified schematic diagram of a GNSS rover process with synthesized base station data according to some embodiments of the invention; Fig. Figure 44 shows a time diagram of a low-latency GNSS rover process with synthesized base station data according to some embodiments of the invention; Fig. Figure 45 shows a time diagram of a high-accuracy GNSS rover process using synthesized base station data according to some embodiments of the invention; Fig. Figure 46 shows a schematic diagram of an alternative GNSS rover process using synthesized base station data according to some embodiments of the invention; Fig. Figure 47 shows the performance of a GNSS rover process with ambiguity setting according to some embodiments of the invention in relation to a GNSS rover process without ambiguity setting or ambiguity setting; Fig. Figure 48 shows a schematic diagram of a GNSS rover process with ambiguity setting according to some embodiments of the invention; Fig. Figure 49 shows a schematic diagram of a GNSS rover process with ambiguity setting according to some embodiments of the invention; Fig. Figure 50 shows a schematic diagram of a GNSS rover process with ambiguity setting according to some embodiments of the invention; Fig. Figure 51 shows a schematic diagram of a GNSS rover process with ambiguity setting according to some embodiments of the invention; Fig. Figure 52 shows a schematic diagram of a GNSS rover process with ambiguity setting according to some embodiments of the invention; Fig. Figure 53 shows a schematic diagram of a GNSS rover process with ambiguity setting according to some embodiments of the invention; Fig. Figure 54 shows a schematic diagram of a GNSS rover process with ambiguity setting according to some embodiments of the invention; Fig. Figure 55 shows a schematic diagram of a GNSS rover process with ambiguity setting according to some embodiments of the invention; Fig. Figure 56 shows a schematic diagram of a GNSS rover process with ambiguity setting according to some embodiments of the invention; Fig. Figure 57 shows a schematic diagram of a GNSS rover process with ambiguity setting according to some embodiments of the invention; Fig. Figure 58 shows a schematic diagram of a GNSS rover process with ambiguity setting according to some embodiments of the invention; Fig. Figure 59 shows a schematic diagram of a GNSS rover process with ambiguity setting according to some embodiments of the invention; Fig. Figure 60 shows a schematic diagram of a GNSS rover process with ambiguity setting according to some embodiments of the invention; Fig. Figure 61 shows a schematic diagram of a GNSS rover process with ambiguity setting according to some embodiments of the invention; Fig. Figure 62 shows a schematic diagram of a GNSS rover process with ambiguity setting according to some embodiments of the invention; DETAILED DESCRIPTION Part 1: System Overview

[0009] Global Navigation Satellite Systems (GNSS) include GPS, Galileo, GLONASS, Compass, and other similar positioning systems. While the examples given here are geared toward GPS processing, the principles are applicable to any such positioning system.

[0010] Definition of Real-Time: The term "real-time" is used several times in this document. Within the scope of the inventions covered by the following embodiments, this term means that an action (for example, data is processed, results are calculated) occurs as soon as the information required for that action is available. Therefore, a certain latency exists, and this depends on various aspects depending on the system component. The information required for the application covered in this document is typically GNSS data and / or GNSS corrections, as described below.

[0011] The network processors, which operate in real time, are capable of providing results for a period of data from a network of monitoring receivers (1a): The data is collected from each of the monitoring receivers (typically less than 1 msec); (1b) the data is transmitted from each receiver to the processing center (typically less than 2 sec); (1c) the data is processed by the processor. The computation of the results by the network processors typically takes between 0.5 and 5 seconds, depending on the processor type and the amount of data to be processed.

[0012] It is normal for data that does not adhere to certain transmission delay restrictions (for example, 3 seconds) to be rejected or buffered and therefore not immediately used for the current epoch update. This prevents an increase in system latency if one or more stations transmit data with an unacceptable amount of delay.

[0013] A rover receiver or mobile unit receiver operating in real time is able to provide results for an epoch of data after the data has been collected by a receiver (typically less than 1 msec) and: (2a) the correction data is generated by the processing center (see 1a, 1b, 1c); (2b) the correction data (if needed) is received from the processing center (typically less than 5 sec); (2c) the data is processed (typically less than 1 msec).

[0014] To avoid or minimize the effect of data latency caused by (2a) and (2b), a delta-phase approach can be used, so that updated receiver positions can be calculated (typically in less than 1 msec) immediately after the data is collected and with correction data streams. The delta-phase approach is described, for example, in US Patent 7,576,690, granted on August 18, 2009, to U. Vollath.

[0015] Fig. 1 and Fig. Figure 2 shows high-level views of a system 100 according to some embodiments of the invention. Reference stations of a worldwide tracking network, such as reference stations 105, 110, ... 115, are distributed around the Earth. The position of each reference station is known very precisely, for example, within less than 2 cm. Each reference station is equipped with an antenna and tracks the GNSS signals transmitted by satellites within line of sight of that station, such as GNSS satellites 120, 125, ... 130. The GNSS signals have codes that are modulated on each of the two or more carrier frequencies. Each reference station acquires GNSS data 205, which for each satellite in sight at each epoch represent carrier-phase (carrier) observations 210 of at least two carriers, and pseudo-distance (code) observations 215 of the corresponding codes, modulated on at least two carriers.The reference stations also receive the satellite almanac (calendar) and ephemerides 220 from the satellite signals. The almanac contains the approximate position of all GNSS satellites, while the so-called transmission ephemerides provide more accurate predictions (approx. 1 m) of the satellites' positions, as well as the time signal error or clock error of the satellites (approx. 1.5 m) over specific time intervals.

[0016] GNSS data collected at the reference stations are transmitted via communication channels 135 to a network processor 140. The network processor 140 uses the GNSS data from the reference stations with other information to generate a correction message containing a precise satellite position and time signal data, as described in detail below. The correction message is transmitted for use by a number of GNSS rover receivers. The correction message is transmitted as shown in Fig. Figure 1 shows a GNSS rover 160 equipped with an uplink 150 and a communications satellite 155 for transmission over a wide area; any other suitable transmission medium may be used, including but not limited to radio transmission or mobile phone connection. The rover 160 is an example of a GNSS rover receiver with a GNSS antenna 165 for receiving and tracking signals from GNSS satellites within line of sight of its position and optionally includes a communications antenna 170. Depending on the transmission band of the correction message, it can be received by the rover 160 via either the GNSS antenna 165 or the communications antenna 170. Part 2: Network Architecture

[0017] Fig. Figure 3 shows a schematic diagram illustrating the main components of the process flow 300 of a network processor 140 according to some embodiments of the invention. GNSS data from the global network of reference stations 310 are fed, without corrections as GNSS data 305 or after correction by an optional data corrector 310 as corrected GNSS data 315, to four processors: a standard time signal processor 320, a Melbourne-Wübbena (MW) bias processor 325, an orbit processor 330, and a phase time signal processor 335.

[0018] The Data Corrector 310 optionally analyzes the raw GNSS data 305 for each reference station to check the quality of the received observations and, if necessary, to correct the data for cycle slips, which are jumps in the carrier-phase observations that occur, for example, whenever the receiver experiences a loss of lock. Commercially available reference stations typically detect cycle slips and mark the data accordingly. Cycle slip detection and correction techniques are summarized, for example, in G. Seeber, SATELLITE GEODESY, 2nd edition (2003), pages 277–281. The Data Corrector 310 optionally applies other corrections. Although not all corrections are required for all processors, applying them to the data does not cause harm.For example, as described below, some processors use a linear combination of code and carrier observations, in which some uncorrected errors are deleted when forming the combinations.

[0019] Observations are recorded epoch-wise at each reference station and transmitted with time tags to the network processor 140. For some stations, the observations arrive late. This delay can range from milliseconds to minutes. Therefore, an optional synchronizer 318 collects the corrected reference station data within a predetermined time span and forwards the observations to the processor as a set for each epoch. This allows data arriving with a reasonable delay to be inserted into an epoch of data.

[0020] The MW bias processor 325 takes either uncorrected GNSS data 305 or corrected GNSS data 315 as input, since it uses the Melbourne-Wübbena linear combination, which eliminates all but the ambiguities and biases of the phase and code observations. Therefore, only receiver and satellite antenna corrections are important for the Widelane processor 325. Based on this linear combination, one MW bias per satellite and one Widelane ambiguity per receiver-satellite pair are calculated. The biases are smoothed (not noisy) and show only some sub-daily variations at a low rate. The Widelane ambiguities are constant and can be used as long as no cycle slippage occurs in the observations at the corresponding satellite-receiver link. Therefore, bias estimation is not very time-critical and can, for example, run with a 15-minute update rate.This is advantageous because the computation time increases with the cube of the number of stations and satellites. For example, the computation time for a reasonable network with 80 stations can be around 15 seconds. The values ​​of the fixed Widelane ambiguities 340 and / or Widelane bias 345 are optionally used in the orbit processor 330 and / or the phase-time signal processor 335, and / or are fed to the scheduler 355. The MW bias processor 325 is described in detail in Part 7 below.

[0021] Some embodiments of the Orbit Processor 330 are based on a prediction-correction strategy. Using a precise force model and starting from an initial assumption of the unknown values ​​of the satellite's parameters (initial position, initial velocity, and dynamic force model parameters), the orbit of each satellite is predicted by integrating the satellite's nonlinear dynamic system. The sensitivity matrix, containing the partial derivatives of the current position with respect to the unknown parameters, is calculated simultaneously. Sensitivities of the initial satellite state are also calculated simultaneously for the entire prediction. This means that the prediction begins with the unknown parameters, the system of differential equations is solved, and the orbit is integrated for the present or future time. This prediction can then be linearized in the direction of the unknown parameters.Therefore, the partial derivatives (sensitivities) serve as a measure of the magnitude of the change in the current satellite states if the unknown parameters are changed, or vice versa.

[0022] In some embodiments, these partial derivations are used in a Kalman filter to improve the initial state estimate by projecting the GNSS observations onto the unknown satellite parameters. This precise initial state estimate is used to reintegrate the satellite's dynamic system and determine a precise orbit. A time update of the initial satellite state to the current epoch is performed periodically. In some embodiments, ionospheric ambiguities are also states of the Kalman filter. The fixed Widelane ambiguity values ​​340 are used to set the ionospheric ambiguities of the orbit processor 330 to improve the accuracy of the estimated orbits. A satellite orbit is very uniform and can be predicted for minutes and hours.These precise orbit predictions 350 are optionally forwarded to the standard time signal processor 320 and to the phase time signal processor 335, as well as to a scheduler 355.

[0023] Ultrafast 360° orbits, such as IGU orbits provided by the International GNSS Service (IGS), can be used as an alternative to precise 355° orbit predictions. IGU orbits are updated four times daily and are available with a three-hour delay.

[0024] The Standard Clock Processor 320 calculates code-matched satellite time signals 360 (also called standard satellite time signals) using GNSS data 305 or corrected GNSS data 315 and using precise orbit predictions 355 or ultrafast orbits 365. Code-matched means that the time signals, or times, are sufficient for use with ionospheric code observations, but not with carrier-phase observations, because the code-matched time signals do not preserve the integer nature of the ambiguities. Code-matched time signals 360 calculated by the Standard Clock Processor 320 represent time signal error differences between satellites.The Standard Time Signal Processor 320 uses the time signal errors of the transmission ephemerides as pseudo-observations and aligns the estimated time signals to a GPS time, allowing them to be used to calculate, for example, the exact time of transmission of a satellite's signal. The time signal errors change rapidly, but for use with code measurements, which are quite noisy, an accuracy of a few centimeters is sufficient. Therefore, a low-rate update rate of 30 to 60 seconds is adequate. This is advantageous because computation time increases with the cube of the number of stations and satellites. The Standard Time Signal Processor 325 also determines tropospheric zenith delays 365 as a byproduct of the estimation process. The tropospheric zenith delays and the code-aligned time signals or times are sent to the Phase Time Signal Processor 335.The Standard Time Signal Processor 320 is described in detail in Part 6 below.

[0025] The phase-time signal processor 335 optionally uses the fixed wide-lane ambiguities 340 and / or MW bias 345 from wide-lane processor 325, together with the tropospheric zenith delays 365 and the precise orbits 350 or IGU orbits 360, to estimate single-differentiated time signal errors or time errors and narrow-lane ambiguities for each pair of satellites. The single-differentiated time signal errors and narrow-lane ambiguities are combined to obtain single-differentiated phase-matched time signal errors 370 for each satellite (except for a reference satellite), which are single-differentiated relative to the reference satellite. The code-matched low-rate time signals 360, the tropospheric zenith delay 365 and the precise orbits 350 or IGU orbits 360 are used to estimate code-matched high-rate time signals 375.Here, the computational effort is linear to the number of stations and to the cube of the number of satellites. For example, the rapidly changing phase-matched time signals 370 and code-matched time signals 375 are available with a delay of 0.1–0.2 seconds. The high-rate phase-matched time signals 370 and the high-rate code-matched time signals 375 are sent to the scheduler 355 along with the MW bias 340. The phase-time signal processor 340 is described in detail in Part 9 below.

[0026] The scheduler 355 receives the orbits (precision orbits 350 or IGU orbits 360), the MW bias 340, the phase-matched high-rate time signals 370, and the code-matched high-rate time signals 375. The scheduler 355 combines these and forwards the combined orbits and time signals, or times and bias 380, to a message encoder 385, which prepares a correction message 390 in a compressed format for transmission to the rover. Transmission to a rover takes approximately 10 to 20 seconds via satellite link, but can also be carried out using a mobile phone, a direct internet connection, or a suitable communication link. Part 3: Observational Data Corrector

[0027] Fig. Figure 4 shows a schematic diagram of a data correction according to some embodiments of the invention. An optional observation corrector 310 corrects the GNSS signals collected at a reference station for station offsets due to centrifugal, gyroscopic, and gravitational forces acting on the Earth, the location of the station's antenna phase center relative to the station's antenna mounting point, the location of the satellite's antenna phase center relative to the satellite's center of mass given by the satellite's orbit, and variations in these phase centers depending on the arrangement of the station antenna and the satellite antenna.

[0028] The main contributors to station offsets are crustal tides up to 500 mm, ocean tidal charges up to 100 mm, and polar tides up to 10 mm. All of these depend on the station's location. A more detailed description is given in McCarthy, DD, Tetit, G. (Editors), IERS Conventions (2003), IERS Technical Note No. 32, and the references cited therein.

[0029] Ocean tides, caused by the forces of astronomical bodies—primarily the Moon—act upon the Earth's loose masses, also causing the Earth's tectonic plates to be lifted and lowered. This well-known effect manifests as recurring variations in the locations of reference stations. Crust tides are optionally calculated for network processing as well as for rover processing, since the effect should not be neglected and the computational effort is small.

[0030] The second largest effect is the deformation of Earth's tectonic plates due to the load of the oceans, which changes over time with the tides. Ocean tide loading parameters, used to quickly calculate the offset of a station over time, depend on the station's location. The computational effort required to derive these parameters is quite high. They can be calculated for a given location using one of the well-known models available from the online Ocean Tide Loading Service provided by the Onsala Space Observatory Ocean, http: / / www.oso.chalmers.se / -loading / , Chalmers: Onsala Space Observatory, 2009. Lower-precision parameters, such as those obtained by interpolating a pre-calculated grid, are sufficient for the applications discussed here.

[0031] The smallest effect mentioned here is that due to polar tides. This offset results from the uplift of a tectonic plate, caused by the centrifugal and gyroscopic effects generated by the Earth's polar motion. Earth orientation parameters are used for this calculation. These are regularly updated by the International Earth Rotation & Reference System Service (IESTS), http: / / hpiers.obspm.fr / , L'Observatoire de Paris 2009, and are not easy to calculate. This small effect is therefore optionally neglected in the rover's processing.

[0032] Absolutely calibrated antenna models are used to calculate the offsets and variations of receiver and satellite antenna phase centers. A description is given by J. Kouba, A Guide to Using International GPS Service (IGS) Products, Geodetic Survey Division, Natural Resources Canada, February 2003. Calibration data collected by the IGS are made available in Antex files at http: / / igscb.jpl.nasa.gov / , 2009; satellite antenna offset information, for example, is given in the IGS absolute antenna file IGS05.atx.

[0033] Another effect is antenna phase wind-up. If a receiver antenna moves relative to the transmitting antenna, the recorded data will show a phase shift. If this effect is neglected, a complete rotation of the satellite around the transmitting axis will cause an error of one cycle in the carrier phase, detected at the receiver. Since the orientation of the satellite relative to the receiver is usually well known, this effect can be represented by a model, as described in Wu JT, Hajj GA, Bertiger WI, & Lichten SM, Effects of antenna orientation on GPS carrier phase, MANUSCRIPTA GEODAETICA, Volume 18, pages 91-98 (1993).

[0034] The relative motion of the station and the satellite is primarily due to the satellite's orbital movement. If a satellite eclipses, this means that its orbit crosses the Earth's shadow, and additional rotations of the satellite around its transmitting axis are possible. For example, GPS Block IIA satellites exhibit a noon rotation and a shadow-crossing maneuver, while GPS Block IIR satellites exhibit a noon rotation and a midnight rotation. If the Sun, Earth, and satellite are nearly colinear, it is difficult to calculate the direction of the rotation maneuver, and an incorrect direction will introduce an error in the carrier phase of a cycle. The satellite's yaw attitude affects the phase wind-up and satellite antenna corrections. Detailed descriptions are provided in Kouba, JAsimplified yaw-attitude model for eclipsing GPS satellites, GPS SOLUTION (2008) and Bar-Server, YE, A new model for GPS yaw attitude, JOURNAL OF GEODESY, volume 70, pages 714-723 (1996).

[0035] In the case of using only phase observations, the effect of an unmodeled satellite rotation maneuver cannot be separated from the satellite time signal or satellite time. Therefore, in a phase-time signal error estimation, the effect of the rotation maneuver is included in the estimated satellite time signal error. If a rover uses these time signals or times, it must not correct for a satellite rotation maneuver.

[0036] The sun's position is needed to calculate the satellite's fixed coordinate system, as the x-axis is defined by the cross product of the satellite's position and the sun's position. This coordinate system is used to calculate the trajectory, satellite antenna correction (offset and variations mapped to the sine of the view), and phase wind-up correction. The moon's position is also required for calculating crustal tides. How the positions of the sun and moon are calculated is shown, for example, in Seidelmann, P.K. (Editor), Explanatory Supplement to the Astronomical Almanac, University Science Books, US (1992).

[0037] Further corrections can also be applied, although these are of less interest to the level of positioning accuracy required by the market.

[0038] Additional effects such as corrections for relativistic effects, ionospheric and tropospheric delays, do not need to be considered in the optional Data Corrector 310. Relativistic corrections are normally applied to the satellite time signals. The largest first-order effect due to the ionosphere is eliminated using an ionospheric-free combination of the GNSS observations, and the effect due to the troposphere is partially modeled and partially estimated in some embodiments. Part 4: Forming Linear Combinations 4.1 Basic Model Equations

[0039] For code Pi,kmj and carrier phase Φi,kmj For observations between receiver i and satellite j on frequency band k and modulation type m, the following observation model is assumed, which relates the observations to certain physical quantities or sizes. Pi,kmj=ρij+cΔti−cΔtj+Tij+IP,i,kj+bP,i,km−bP,kmj+mP,i,kmj+εP,i,kmj. Φi,kmj=ρij+cΔti−cΔtj+Tij+IΦ,i,jj+bΦ,i,j−bΦ,kj+λkNi,kj+mΦ,i,kmj+εΦ,i,kmj. with geometric distance from satellite j to receiver i c Speed ​​of light Δt i Receiver i-time signal error Δt j Satellite j-time signal error tropospheric delay from satellite j to receiver i code-ionospheric delay at frequency f k Carrier-phase ionospheric delay on f k b P,i,km Code receiver bias Code satellite bias b Φ,i,k Phase receiver bias (independent of modulation type m) Phase satellite bias (independent of modulation type m) integer ambiguity expression from satellite j to receiver i on λ k Code multipath from satellite j to receiver i Phase multipath from satellite j to receiver i Code random noise expression Phase random noise expression

[0040] The modulation type dependency in phase observation can be suppressed by assuming that the different phase signals on a frequency band are already shifted to a common basis within the receiver. For example, L2C is assumed to be shifted by -0.25 cycles to compensate for the 90-degree rotation of the quadrature phase on which it is modulated. However, noise multipath expressions (which are not usually represented by a model) still have a differential contribution to phase observation for different modulation types.

[0041] Examples of the different modulation types (also called code types) are, in the case of GPS, L1C / A, L1P, L1C on an L1 frequency band and L2C, L2P on an L2 frequency band, and in the case of GLONASS, L1C / A, L1P and L2C / A, L2P. For the GLONASS satellite system, the wavelength λ depends k and frequency f k also depends on a satellite-specific frequency channel number, so the notation could be extended to λk(j) and ƒk(j). Additionally, the code receiver bias b also has a special effect. P,i,km also a channel and therefore a satellite dependency (as can be seen in a zero-baseline processing with a little averaging over time, so that Pi1i2,kmj=bP,i1,i2,km). Therefore, a more precise formulation for the code-receiver bias would be bP,i,kmj=bP,j,km+ΔbP,i,kmj.

[0042] The symbol Φ, used here for carrier-phase observations, is also used for the time-transition matrix in the Kalman filter context. In both cases, Φ is the standard symbol used in scientific literature, and we adopt this notation. The meaning of Φ will always be clear from the context.

[0043] In the following, we will discuss the second order I2ijƒk3 and the third order I3ijƒk4 neglect ionospheric expressions, which are typically in the range of less than 2 cm (Morton, van Graas, Zhou, & Herdtner, 2008), (Datta-Barua, Walter, Blanch, & Enge, 2007). In this way, it applies IP,i,kj=Iijƒk2=−IΦ,i,kj with Iij:=I1ij. Only under very strict geomagnetic active conditions can the second- and third-order expressions reach several tens of centimeters. However, these conditions only occur for a few days every many years. Higher-order ionospheric expressions can be considered using ionospheric models based on the Appleton-Hartree equation, which determines the phase index of the refraction of a right-handed circularly polarized wave propagation through the ionosphere at the wave frequency f. k , relates the electron density and the Earth's magnetic field. Approximations to the Appleton-Hartree equation allow the parameters to be approximated. I2ij,I3ij the second and third order ionospheric expressions to ionospheric estimation parameters Iij:=I1ij to relate to the first order, which is a measure of the total electron content along the signal propagation path. Therefore, higher-order ionospheric expressions can be corrected based on observational data at at least two frequencies.

[0044] In the following, we will often refer to ionospheric-free (IF) linear combinations. However, it should be noted that these linear combinations only eliminate the first-order ionospheric expression and are therefore not completely ionospheric-free. Linear combinations of observations

[0045] By combining multiple codes Pi,kmj and inert phases Φi,kmj Observations in a linear way LC=∑i,j,k,maP,i,kmjPi,kmj+aΦ,i,kmjΦi,kmjwith aP,i,kmj,aΦ,i,kmj∈R for all i,j,k,m Some physical quantities can be eliminated by the linear combination LC, so these quantities do not need to be estimated if the linear combination is used as the observation input for an estimator. In this way, some linear combinations are of particular importance.

[0046] Single difference (SD) observations between two satellites j1 and j2 eliminate all quantities that are not satellite-dependent, i.e., that do not have a satellite index j.

[0047] Under definition X j1j2 := X j2 - X j1 The intersatellite SD observations are formally obtained by substituting each index j with j1 and j2 and ignoring all expressions without a satellite index j. Pi,kmj1j2=ρij1j2−cΔtj1j2+Tij1j2+JP,i,kj1j2−bP,kmj1j2+mP,i,kmj1j2+εP,i,kmj1j2. Φi,kmj1j2=ρij1j2−cΔtj1j2+Tij1j2+IΦ,i,kj1j2−IΦ,kj1j2+λkNi,kj1j2+mΦ,i,kj1j2+εΦ,i,kmj1j2.

[0048] In this way, the receiver time signal and receiver bias expressions were eliminated in the linear combination.

[0049] In the same way, single difference observations between 2 receivers i1 and i2 eliminate all quantities that are not receiver-dependent, that is, that do not have a receiver index i.

[0050] By generating the difference between two receivers i1 and i2 with respect to the intersatellite single-difference observations (4) and (5), double-difference (DD) observations are obtained, which also eliminate all receiver-dependent expressions of (4) and (5).

[0051] By defining Xi1i2j1j2:=Xi2j1j2−Xi1j1j2=(Xi2j2−Xi2j1)−(Xi1j2−Xi1j1) The DD observations are formally obtained from (4) and (5) by substituting each index i with i1i2 and ignoring all expressions without a receiver index i. Pi1i2,kmj1j2=ρi1i2j1j2+Ti1i2j1j2+IP,i1i2,kj1j2+mP,i1i2,kmj1j2+εP,i1i2,kmj1j2. Φi1i2,kmj1j2=ρi1i2j1j2+Ti1i2j1j2+IΦ,i1i2,kj1j2+λkNi1i2,kj1j2+MΦ,i1i2,kmj1j2+εΦ,i1i2,kmj1j2.

[0052] In this way, the satellite time signals and satellite bias were also eliminated in the linear combination.

[0053] In the following, we assume that all code observations Pi,kmj correspond to the same modulation type and that all phase observations Φi,kmj correspond to the same observation type, which may differ from the modulation type of the code observations. Since the modulation type dependency for the phase observations only occurs in the multipath and random noise expressions not represented as a model, the modulation type index m can be suppressed in this way.

[0054] For our purposes, two linear combinations that introduce a first-order ionospheric delay are suitable. Iijƒk2 The different methods of annihilation are of particular importance, including the iono-free linear combination for code and carrier phase and the Melbourne-Wübbena (MW) linear combination. Φi,WLj−Pi,NLj consisting of the Widelane (WL) carrier phase Φi,WLjλWL:=Φi,1jλ1−Φi,2jλ2 and Narrowlane (NL) code Pi,NLjλNL:=Pi,1jλ1−Pi,2jλ2 Observations with wavelengths λWL=cƒWL:=cf1−f2 and λNL=cƒNL:=cƒ1+ƒ2, (Melbourne, 1985), (Wübbena, 1985), Pi,NLj=ρij+cΔti−cΔtj+Tij+λNL(bP,i,1λ1+bP,i,2λ2)−λNL(bP,1jλ1+bP,2jλ 2)+λNL(Iijλ1ƒ12+Iijλ2ƒ22)+λNL(mP,i,1j+εP,i,1jλ1+mP,i,2j+εP,i,2jλ2) Φi,WLj=ρij+cΔti−cΔtj+Tij+λWL(bΦ,i,1λ1+bΦ,i,2λ2)−λNL(bΦ,1jλ1+bΦ,2jλ2)+λWL(Iijλ1 ƒ12−Iijλ2ƒ22)+λWL(Ni1j−Ni2j)︸=:Ni,WLj+λWL(mΦ,i,1j+εΦ,i,1jλ1+mΦ,i,2j+εΦ,i,2jλ2) so dass Φi,WLj−Pi,NLj=λWL(bΦ,i,1λ1−bΦ,i,2λ2)−λNL(bP,i,1λ1+bP,i,2λ2)︸=:bi,MW+−[λ WL(bΦ,1jλ1−bΦ,2jλ2)−λNL(bP,1jλ1+bP,2jλ2)]︸=:bMWj+λWLNi,WLj++λWL(mΦ,i,1jλ 1−mΦ,i,2jλ2)−λNL(mP,i,1jλ1+mbP,i,2jλ2)︸=:mi,MWj++λWL(εΦ,i,1jλ1−εΦ,i,2jλ2 )−λNL(εP,i,1jλ1+mεP,i,2jλ2)︸=:εi,MWj+=bi,MW−bMWj+λWLNi,WLj+mi,MWj+εi,MWj where the ionospheric output in the WL phase is solved with the ionospheric output in the NL code underlying −λWL(1λ1ƒ12−1λ2ƒ12)−λNL(1λ1ƒ12+1λ2ƒ12)=−cƒ1−ƒ2(cƒ1−cƒ2)−cƒ1+ƒ 2(cƒ1+cƒ2)=−c2ƒ1−ƒ2ƒ2−ƒ1ƒ1ƒ2−c2ƒ1+ƒ2ƒ2+ƒ1ƒ1ƒ2=+c2ƒ1ƒ2−c2ƒ1ƒ2=0

[0055] Neglecting the ordinary not as a model there multifaceted I,MWj and Random Noise Pressure εi,MWj simplifies Equation (10) to Φi,WLj−Pi,NLj=bi,MW−bMWj+λWLNi,WLj or in an inter-satellite single difference (SD) version Φi,WLj1j2−Pi,NLj1j2=−bMWj1j2+λWLNi,WLj1j2

[0056] It should be noted that the satellite bias cancels out in the double-difference (DD) (between receivers and between satellites) Melbourne-Wübbena (MW) observation. Φi1i2,WLj1j2−Pi1i2,NLj1j2=λWLNi1i2,WLj1j2.

[0057] Therefore, the DD-WL ambiguities Ni1i2,WLj1j2 directly observed through the DD-MW observations.

[0058] The iono-free linear combination with respect to code Pi,IFj=ƒ12Pi,1j−ƒ22Pi,2jƒ12−ƒ22 and carrier phase Φi,IFj:=ƒ12Φi,1j−ƒ22Φi,2jƒ12−ƒ22 results Pi,IFj=ρij+cΔti−cΔtj+Tij+ƒ12bP,i,1−ƒ22bP,i,2ƒ12−ƒ22︸=:bP,i,IF−ƒ12bP,1j−ƒ22bP,2jƒ12−ƒ22︸=:bP,IFj+ƒ12mP,ji,1−ƒ22mP, i,2jƒ12−ƒ22︸=:mP,i,IFj+ƒ12εP,i,1j−ƒ22εP,i,2jƒ12−ƒ22︸=:εP,i,IFj=ρij+cΔti−cΔtj+Tij+bP,i,IF−bP,IFj+mP,i,IFj+εP,i,IFj und Φi,IFj=ρij+cΔti−cΔtj+Tij+ƒ12bΦ,i,1−ƒ22bΦ,i,2ƒ12−ƒ22︸=:bΦ,i,IF−ƒ12bΦ, 1−ƒ22bP,2jƒ12−ƒ22︸=:bΦ,IF+ƒ12λ1Ni,1j−ƒ22λ2Ni,2jƒ12−ƒ22︸=:λIFNi,IFj+ƒ1 2mΦ,i,1j−ƒ22mΦ,i,2jƒ12−ƒ22︸=:mΦ,i,IFj+ƒ12εΦ,i,1j−ƒ22εΦ,i,2jƒ12−ƒ22︸=: εΦ,i,IFj=ρij+cΔti−cΔtj+Tij+bΦ,i,IF−bΦ,IFj+λIFNi,IFj+mΦ,i,IFj+εΦ,i,IFj

[0059] Apparently it is usually not as good as the model labeled Multiphads mP,i,IFj,mΦ,i,IFj und Zufallsrauschenaudrücke εP,i,IFj,εΦ,i,IFj, refine (14) and (15) to Pi,IFj=ρij+cΔti−cΔtj+Tij+bP,i,IF−bP,IFj. Φi,IFj=ρij+cΔti−cΔtj+Tij+bΦ,i,IF−bΦ,i,IFj+λIFNi,IFj. or in an inter-satellite-individual difference- (SD) - version to Pi,lFj1j2=ρii1j2−cΔtj1j2+Tij1j2−bP,IFj1j2. Φi,IFj1j2=ρij1j2−cΔtj1j2+Tij1j2−bΦ,IFj1j2+λIFNi,IFj1j2.

[0060] The ion-free wavelength λIF depends only on the ratio of the frequencies involved, which are listed for different GNSS systems in Table 1 and Table 2. Table 1 GPS Galileo L1 L2 L5 E2L1E1 E5a E5b E5ab E6 10,23 MHz 154 120 115 154 115 118 116,5 125 Table 2 Glonass (k = -7,...,+6) L1L2=97 L1 L2 (1602 + k · 9 / 16) MHz (1246+k·7 / 16) MHz

[0061] Defining F1, F2 ∈ N by ƒ1=:F1gcd(ƒ1,ƒ2)ƒ2=:F2gcd(ƒ1,ƒ2)⇒ƒ1ƒ2=F1F2 where gcd is an abbreviation for greatest common divisor, it follows for the iono-free wavelength λIFNi,IFj:=ƒ12λ1Ni,1j−ƒ22λ2Ni,2jƒ12−ƒ22=λ1ƒ12Ni,1j−ƒ22λ2λ1Ni ,2jƒ12−ƒ22=λ1ƒ12Ni,1j−ƒ22ƒ1ƒ2Ni,2jƒ12−ƒ22=λ1ƒ12ƒ22Ni,1j−ƒ1ƒ2N i,2jƒ12ƒ22−1=λ1F12F22Ni,1j−F1F2Ni,2jF12F22−1=λ1F1F22F1Ni,1j− F2Ni,2jF12−F22F22=λ1F1F12−F22︸=:λIF(F1Ni,1j−F2Ni,2j)︸:=Ni,IFj

[0062] The factors F1 and F2 are listed for different GNSS frequency combinations along with the resulting ion-free wavelengths in Table 3. Table 3 GNSS Freq. bänder F1 / F2 λ1 / m l NL / m λIFλ1=F1F12−F22 GPS L1-L2 77 / 60 0,1903 0,1070 0,0331 LI-L5 154 / 115 0,1903 0,1089 0,0147 L2-L5 24 / 23 0,2442 0,1247 0,5106 Galileo L1-E5a 154 / 115 0,1503 0,1089 0,0147 L1-E5b 77 / 59 0,1903 0,1077 0,0315 L1-E6 154 / 125 0,1903 0,1050 0,019 E5b-E5a 118 / 115 0,2483 0,1258 0,1688 E6-E5a 25 / 23 0,2344 0,1221 0,2604 E6-E5b 125 / 118 0,2344 0,1206 0,0735 Glonass L1-L2 9 / 7 c(1602+k⋅9 / 16)⋅106 c(2848+k)⋅106 0,2813

[0063] Since the ion-free wavelength λ is used for most frequency combinations IF Since the frequency combination L2-L5 is too small for reliable ambiguity resolution (it is a notable exception), the following relationship exists between the iono-free ambiguity. Ni,IFj and the Widelange ambiguity Ni,WLj of particular importance. By using the definitions Ni,WLj:=Ni,1j−Ni,2jNi,NLj:=Ni,1j+Ni,2j}⇔Ni,1j=12(Ni,NLj+Ni,WLj)Ni,2j=12(Ni,NLj−Ni,WLj) The iono-free ambiguity expression can be rewritten as λIFNi,IFj:=ƒ12ƒ12−ƒ22λ1Ni,1j−ƒ22ƒ12−ƒ22λ2Ni,2j=(ƒ12ƒ12−ƒ22cƒ 1+ƒ22ƒ12−ƒ22cƒ2)︸cƒ1(ƒ1−ƒ2)(ƒ1+ƒ2)+cƒ2(ƒ1−ƒ2)(ƒ1+ƒ2)=cƒ1−ƒ2=: λWL12Ni,WLj+(ƒ12ƒ12−ƒ22cƒ1−ƒ22ƒ12−ƒ22cƒ2)︸cƒ1(ƒ1−ƒ2)(ƒ1+ƒ2)−c ƒ2(ƒ1−ƒ2)(ƒ1+ƒ2)=cƒ1+ƒ2=:λNL12Ni,NLj=12λNLNi,NLj+12λWLNi,WLj; Ni,NLj=Ni,1j+Ni,2j=Ni,1j+12(Ni,NLj−Ni,WLj)=λNLNi,1j+12(λWL−λNL)Ni,WLj

[0064] Therefore you can, because of the Widelane-Mehrdeutigkeit Ni,WLj was celebrated at a considerable cost based on the Melbourne-Wübbena Line combination (11), the zone (24) being used for a considerable outflow of the unrestricted Narrowlane Meadow Ni,1j (especially if λ NL >> λ IF , see Table 3), Ni,1j=1λNL(λIFNi,IFj−12(λWL−λNL)Ni,WLj)

[0065] We are together Ni,1j the unbounded or free narrowlane-mean values, since they are derived in (24) in combination with the narrowlane-well length λNL and does not depend on whether the designated wide lane is even or odd. Since N NL = N WL + 2N2 (see (23)), has N NL always, for consistency reasons, the same even / odd status as N WL and is therefore already limited to a certain extent. Part 5: Kalman Filter Overview

[0066] Some embodiments of the Standard Time Signal Processor 320, the MW Bias Processor 325, the Orbit Processor 230 and the Phase Time Signal Processor 335 use a Kalman filter approach.

[0067] Fig. Figure 6 shows a diagram of the Kalman filter algorithm 600. The modeling, which is subject to the Kalman filter equations, relates the state vector x. k at time step k (containing the unknown parameters) on the observations (measurements) z k via the design matrix H kand to the state vector at time step k-1 via the state transition matrix Φ k-1 together with process noise w k-1 and observational noise v k , whose covariance matrices are assumed to be Q k , R k are known accordingly. Then the Kalman filter equations give the estimated state x̂. k-1 ahead together with the covariance matrix P k-1 via the state transition matrix Φ k-1 and process noise input, described by the covariance matrix Q k-1 at time step k, which is in a predicted state x^k− and predicted covariance matrix Pk− The predicted state matrix and state covariance matrix are then corrected by the observation z. k in the Kalman filter measurement update, where the gain matrix K kplays a central role in state updating, as well as in state covariance updating. Part 6: Code-adapted time signal processor

[0068] The estimated absolute code-adjusted satellite time signals, or low-rate 365 satellite times, are used in positioning solutions, for example, to calculate the precise transmission time of the GNSS signal and also to achieve fast convergence of float position solutions, such as in precise point positioning. For transmission time calculations, an approximate but absolute satellite time signal error estimate can be used. Even the satellite time signals or satellite times from the transmission message are good enough for this purpose. However, the quality of the individually differentiated pairs of satellite time signal errors is important for achieving fast convergence in positioning applications. For this purpose, a time signal accuracy level of approximately 10 cm is desired.

[0069] Some embodiments use quality-controlled ionospheric-free combinations of GNSS observations from a global tracking network, optionally corrected for known effects, to estimate (mostly) continuous absolute code-matched satellite time signal errors.

[0070] The raw GNSS reference station data 305 are optionally corrected by a data corrector 310 to obtain corrected network data 315, as described in Part 3 above.

[0071] For each station, ionospheric-free combinations derived from an observation of signals with different wavelengths (for example, L1 and L2) and the transmitted time signal error predictions are used as an input for the filter: Pr,IFs−cΔtrels=ρrs+cΔtP,r−cΔtPs+Trs+εP,r,IFs Φr,IFs−cΔtrels=ρrs+cΔtP,r−cΔtPs+Trs+λNrs+εΦ,r,IFs Δtbrcs≈ΔtPs where is the ionospheric-free code combination for each receiver-satellite pair r,s is the ionospheric-free phase observation for each receiver-satellite pair r,s is the transmission satellite time signal error prediction is the geometric distance from satellite s to receiver r represents the relativistic effects for satellites cΔt P,r :=cΔt r +b P,r,IF is the time signal error for receivers r is the time signal error for satellite s is the tropospheric delay observed at receiver r represents noise in code measurement represents noise in carrier measurement is the float carrier ambiguity from satellite s to receiver r

[0072] The code and carrier observations are corrected for relativistic effects. Δtrels, This expression is calculated based on satellite orbits when the satellite time signal error is estimated. It can then be added to the estimated time signal error to allow the rover using these satellites to correct for all related effects at once.

[0073] The geometric distance ρrs For each epoch, calculations can be performed using a precise satellite orbit and a precise reference station location. The corresponding noise expressions εP,r,IFs and εΦ,r,IFs The observations for code and carrier are not the same. Differentiating directly between phase observation and code observation leads to an approximate estimate of carrier ambiguity. Nos, although influenced by measurement noise εP,r,IFs and εΦ,r,IFs: Φr,IFs−Pr,IFs=λNrs+εΦ,r,IFs+εP,r,IFs.

[0074] Therefore, this difference is used as an input for the filter. Φr,IFs−Pr,IFs, the phase measurement Φr,IFs and the transmitted satellite time signal error prediction Δtbrcs used. This difference Φr,IFs−Pr,IFs This is a pseudo-measurement for the ambiguities, which are represented as constants through modeling. Since, due to the bias, the floating ambiguity (or floating-number ambiguity) is not truly constant, the estimated value represents the ambiguity along with a constant portion of the bias. The non-constant portion of the bias will be reflected in the residual values. This approximation yields acceptable results as long as the biases are more or less constant. The converged floating ambiguities are used to define the level of the time signal errors.

[0075] Once the ambiguities have converged, the phase measurement Φr,IFs A measurement for time signal errors and the troposphere is available. For the troposphere Trs=(1+cT,r)T^rs Is it sufficient to use only one scaling factor per receiver c? T,r to estimate. A mapping to different altitudes is calculated using a tropospheric model. T^rs. The scaling factor can be assumed to vary over time like a random walk process.

[0076] A linear-time discrete process is assumed for the satellite time signals. ΔtPs(ti+1)=ΔtPs(ti)+w1s(ti)+(Δt˙Ps(ti)+w2s(ti))(ti+1−ti) with the random walks w1s and w2s superimposed on the time signal error ΔtPs and the time signal error rate ΔtPs. Receiver time signals are usually not as precise as satellite time signals and are often unpredictable. Therefore, receiver time signals are represented by a model as white noise to indicate any behavior they might exhibit.

[0077] The system of receiver and satellite time signals is indeterminate if only code and phase observations are used. Therefore, all time signal estimates can exhibit a common trend (an arbitrary function added to each of the time signals or clocks). In a single difference, this trend cancels out, and each single-difference time signal is correct. To overcome this lack of information, the transmission time signal error predictions can be used as pseudo-observations for the satellite time signal errors to keep the system close to GPS time.

[0078] Assuming a random walk with respect to the time signal rate is equivalent to assuming a random run with respect to the time signal error itself. Optionally, a quadratic time signal error model is used to represent changes in the time signal rate. This additional parameter for time signal rate changes can also be represented as a random walk through a model. The noise input used for the filter can be derived by analyzing the time signal errors using, for example, the (modified) Allan deviation or the Hadamard variance. The Allan deviation is described in A. van Dierendonck, "Relationship between Allan Variances and Kalman Filter Parameters," PROCEEDINGS OF THE 16 THANNUAL PRECISE TIME AND TIME INTERVAL (PTTI) SYSTEMS AND APPLICATION MEETING 1984, Seiten 273-292. Die Hadamard-Varianz wird beschrieben in S. Hutsell, Relating the Hadamard variance to MCS Kalman filter clock estimation, 27 TH ANNUAL PRECISE TIME AND TIME INTERVAL (PTTI) APPLICATIONS AND PLANNING MEETING 1996, Seiten 291-301.

[0079] There are many different approaches to dealing with the underdetermined time signal system or clock system, besides adding the transmitted satellite time signal errors as pseudo-observations. One is to fix one of the satellite or receiver time signal errors to the values ​​of an arbitrarily chosen function (for example, 0 or additional measurements of a good receiver time signal). Another is to fix the mean of all time signals to a single value, for example, the mean of transmitted or ultrafast time signal errors, as is done in A. Hausschild, Real-time Clock Estimation for Precise Orbit Determination of LEO-Satellites, ION GNSS 2008, Sept. 16-19, 2008, Savannah, Georgia, pp. 9 ff. This is taken into account when deriving the time signal models or...Clock models; the system model and the noise model fix the time signal error difference to the fixed time signal error and no longer to the original time signal error.

[0080] Fig. Figure 7A shows a schematic diagram of a "standard" code-matched time signal processor 320 according to some embodiments of the invention. An iterative filter, such as a Kalman filter 705, uses, for example, ionospheric-free linear combinations 710 of the reference station observations and time signal error models 715 with transmission satellite time signals 720 as pseudo-observations to estimate code-matched (standard) low-rate satellite time signals 365, tropospheric delays 370, receiver time signals 725, satellite time signal rates 730, (optionally) ionospheric-free sliding ambiguities 374, and (optionally) ionospheric-free code carrier biases 372.

[0081] Further improvements can be made to the quality of the time signals or clocks. Individual differences in the estimated time signal errors can indicate a slow drift due to residual errors in the corrected observations, errors in the orbits, and long-term bias drift. After some time, the individual differences in the estimated time signal errors no longer correspond to a code-matched time signal or clock. To account for such drift, the discrepancy between code and phase measurements is optionally estimated and applied to the estimated time signal errors. In some embodiments, this is done by setting up an additional filter, such as filter 735 from [reference missing]. Fig. 7A, with only one bias per satellite and one per receiver, to estimate the ionospheric-free code carrier biases 372, as designated by "Option 1". The receiver biases are represented as a model, for example, as processes of white noise. The satellite biases are modeled, or represented as a model, for example, as a random walk with a suitable small input noise, because only low-rate variations of the satellite bias are expected. Observations used for this filter include, for example, an ionospheric-free code combination 740, reduced by the tropospheric delay 370, the satellite time signal errors 365, and the receiver time signal errors 725, estimated in the standard code-matched time signal filter 705 above.Instead of setting up the additional filter, such as filter 730, the iono-free code carrier biases are modeled in some embodiments as additional states in the code-adapted time signal estimator filter 705, as indicated by “Option 2”. Part 7: MW (Melbourne-Wübbena) Bias Processor Part 7.1 MW Bias: Motivation

[0082] The distance signals emitted by navigation satellites and received by GNSS receivers contain a component due to delays in the satellite hardware. These hardware delays are commonly called satellite biases or uncalibrated delays. In differential GNSS processing, satellite biases are irrelevant if both receivers receive the same code signals (for example, in the case of GPS, both L1C / A or both L1P). However, biases are always important for Precise Point Positioning (PPP) applications, where the precise positioning of a single rover receiver is achieved using precise satellite time signals.Satellite time signals and precise orbits, determined based on a global network of reference stations (such as the International Global Navigation Service (IGS)) (Zumberge, Heflin, Jefferson, Watkins, & Webb, 1997), (Héroux & Kouba, 2001). Here, knowledge of satellite biases can enable the resolution of undifferentiated (differentiated between satellites) integer ambiguities from the rover, which is key to fast, high-precision positioning without a reference station (Mervart, Lukes, Rocken, & Iwabuchi, 2008), (Collins, Lahaye, Héroux, & Bisnath, 2008).

[0083] Satellite biases are usually assumed to be nearly constant over periods of weeks (Ge, Gendt, Rothacher, Shi, & Liu, 2008), and their variations can be neglected (Laurichesse & Mercier, 2007), (Laurichesse, Mercier, Berthias, & Bijac, 2008). Our own intensive studies, by processing GPS data from a global network of reference stations over several months in the manner proposed here, showed that there are daily repeating patterns in the Melbourne-Wübbena (MW) linear combination of satellite biases of sizes up to approximately 14 cm over 6 hours, as well as drifts over a month of up to approximately 17 cm, and sometimes sudden bias level changes (of arbitrary size) from individual satellites within seconds (for example, GPS PRN 24 on 2008.06.26). Nevertheless, the daily repeatability of the MW satellite bias is usually in the range of 2 to 3 cm, which is consistent with the literature.Therefore, the real-time estimation of satellite biases as a dynamic system in a sequential least-squares filter (such as a Kalman filter ((Grewal & Andrews, 2001), (Bierman, 1977)) and the transmission of these biases to PPP-based rover receivers (in addition to precise satellite time signals and orbits) is important for integer ambiguity resolution at the rover. Part 7.2 MW Bias: Process Flow

[0084] Fig. Figure 11 shows a schematic diagram of a process flow 1100 for MW satellite bias and WL ambiguity determination according to some embodiments. GNSS observation data 305 regarding code and carrier phases on at least two frequencies from a number of reference station receivers are used as the main input to the process. These observations are optionally corrected at 310 for effects that do not require an estimation of any model parameters. Among the corrections typically used in PPP applications for the MW linear combination, receiver and satellite antenna offsets and variation are of particular importance. Higher-order ionospheric expressions do not cancel each other out in this linear combination. The optionally corrected GNSS observation data 315 are then passed to a module 1105, which generates linear combinations 1110 of the code and phase observations on two frequencies.The specified MW observation combinations 1110 are then entered into a sequential filter 1115, such as a Kalman filter), which contains the MW observations. Φi,WLj−Pi,NLj relates to the estimation parameters, that is, the MW satellite biases bMWj 1120, WL ambiguities Ni,WLj 1125 and optional MW receiver bias b i,MW 1130 via equation (11) in the non-differentiated case or via equation (12) in the inter-satellite single-difference case.

[0085] It is important that a process noise input with regard to the MW satellite biases bMWj This ensures that the biases can vary over time. Due to the periodic behavior of the satellite biases, the biases can optionally also be represented as a model using harmonic functions, for example as bMWj=b0j+b1jsin(αj)+b2jcos(αj) where α jthe position of satellite j in the orbit (for example α) j (could be the argument of latitude or the true anomaly) defined and b0j,b1j,b2j The estimated parameters require much less process noise than a single parameter. bMWj and can therefore further stabilize the estimation process.

[0086] In addition to the observational data, a single mean-value bias condition 1140 and several ambiguity conditions 1145 are entered into the filter

[935] . These conditions, or boundary conditions, are additional arbitrary equations that, for example, take the following form: bWLj0=0 Ni,WLj=round(Ni,WLj) and which are added to the filter with a very low observational variance of, for example, 10 -30 m 2The conditions ensure that the system of linear equations in filter 1115 is not underdetermined, so that the variances of the model parameters are immediately of the same order of magnitude as the observational variance. They must be chosen carefully so that the system of linear equations is not overrestricted by restricting a double-difference white matter ambiguity directly given by the mean-value observations (see equation 13). By restricting the ambiguities to an arbitrary integer, information about the integer nature of the ambiguities is introduced into the system. In a Kalman filter approach, where the system of equations in (11) or (12) (optionally together with (31)) is extended by arbitrary or , respectively,For arbitrary equations for the initial values ​​of all parameters, such that a well-defined floating solution (with variances of the size of the initial variances) always exists, it is preferred to restrict the ambiguities to the nearest integer of the Kalman filter floating solution.

[0087] The estimated MW satellite biases 1120 are used either directly as the process output or after an optional additional WL ambiguity-setting step. Therefore, the estimated WL ambiguities 1125 are optionally fed into an ambiguity-setting module 1135. The resulting fixed WL ambiguities 340 (which are either fixed to integers or floating-point values) are used as the second process output. Optionally, the fixed WL ambiguities 340 are fed back into the filter 1115 (or into a filter copy or a secondary filter without ambiguity states (compare Fig. 24A - Fig. 25D) to obtain a satellite MW bias output of 1120 that is consistent with integer WL ambiguities.

[0088] The MW satellite biases 1120, for example, are transferred via the scheduler 355 to rover receivers, where they assist in establishing WL ambiguities at the rover. Network WL ambiguities 1125 or 340 can be forwarded to the phase-time signal processor 335 and orbit processor 330, where they assist in establishing iono-free (IF) ambiguities when the reference station data 305 from the same receiver network is used in these processors. Alternatively, instead of the network WL ambiguities 1125 or 340, MW satellite biases 1120 are transferred to the orbit processor 330 and phase-time signal processor 335 to derive WL ambiguities in a station-wise process for the network receivers in the same manner as is done at the rover. Then the derived WL ambiguities help in defining IF ambiguities.With such an approach, GNSS data from different reference station networks can be used in the different processors.

[0089] Fig. Figure 12 shows a schematic diagram of a processing architecture 1200 according to some embodiments. Code and carrier phase observations 1205 (for example, from reference station data 305) on at least two frequencies from a number of reference station receivers are input into a linear combiner 1210, which generates a set of Melbourne-Wübbena (MW) linear combinations 1220, one such WM combination for each station-satellite pairing of code and carrier phase observations on two frequencies. If more than two frequencies are available, several MW combinations can be generated for a single station-satellite pairing.These MW observations are then fed into a processor 1225, which estimates at least MW biases per satellite 1230, as well as WL ambiguities per station-satellite pair 1235, based on modeling equations (11) in the undifferentiated case or (12) in the inter-satellite single-difference case (both optionally together with (31)). The processor is usually one or more sequential filters, such as one or more Kalman filters. Since it can also consist of multiple filters, the more general term "processor" is used here. Process noise 1240, which is fed onto the MW satellite biases in the processor, allows for variation from epoch to epoch, even after the convergence phase of a filtering. The outputs of the process are the estimated satellite MW biases 1230 and network WL ambiguities 1235.

[0090] Therefore, some embodiments provide a method of processing a set of GNSS signal data derived from code observations and carrier phase observations at multiple receivers of GNSS signals from multiple satellites over multiple epochs, wherein the GNSS signals have at least two carrier frequencies, comprising: forming an MW (Melbourne-Wübbena) combination per receiver-satellite pair at each epoch to obtain one MW data set per epoch;and estimating from the MW data set per epoch an MW bias per satellite, which may vary from epoch to epoch, and a set of WL (widelane) ambiguities, where each WL ambiguity corresponds to a receiver-satellite link and a satellite-receiver-satellite link, where the MW bias per satellite is represented by a model as one of (i) a single estimation parameter and (ii) an estimated offset plus harmonic variations with estimated amplitudes.

[0091] The transmission satellite orbits 1245, included in the navigation message, are used optionally, for example, for coarse receiver positions 1250, to reduce the incoming observations to a minimum elevation angle at which a satellite is seen from a station. The receiver positions 1250 are optionally output as an additional input or can alternatively be derived, as is known in engineering, from the code observations and the transmission satellite orbit. The restriction to observations of a minimum elevation or altitude can be performed at any position before the observations are input into the processor. However, performing the restriction immediately after the code and carrier-phase observations are introduced into the process prevents unnecessary calculations.

[0092] Fig. 13A and Fig. Figure 13B shows the state vectors for the undifferentiated (= zero-differentiated (ZD)) and single-differentiated embodiments, listing the parameters to be estimated. The ZD state vector 1310 comprises n satellite bias states, bMWj, a number of ambiguity states Ni,WLj, which change over time with the number of satellites visible to the stations, and in receiver bias states b i,MW The SD state vector 1320 comprises n satellite bias states. bMWj0 to a fixed reference satellite j0, which can be either a physical or artificial satellite. Additionally, the SD state vector includes the same number of ambiguity states as in the ZD case. However, here each ambiguity represents an SD ambiguity with respect to either a physical or artificial satellite. Each station can have its own choice of reference satellite. In the SD case, no receiver bias states are needed, so there are always m fewer states in the SD state vector 1320 than in the comparable ZD state vector 1310. Further details about artificial reference satellites follow in Part 7.4. Part 7.3 MW Process: Correction and Smoothing

[0093] Fig. Figure 14 shows a process 1400 with the addition of an observation correction to the MW process of Fig. 12. Some embodiments add the observation data correction module 310 from Fig. 3. Code and carrier phase observations 1205 on at least two frequencies from a number of reference stations (for example, from reference station data 305) are corrected in the optional observation data corrector 310 for effects that do not require an estimate of any model parameter (specifically receiver and satellite antenna offsets and variations and higher-order ionospheric effects). The knowledge of the transmission satellite orbits 1245 and rough receiver positions 1250 is used for this. The corrected observation data 1310 are then optionally fed into the process of Fig. 12 for creating MW satellite biases 1330 and wideline ambiguities 1335.

[0094] in Fig. 15. Code and carrier phase observations 1205 on at least two frequencies from a number of reference stations (for example, from reference station data 305) are optionally corrected in the observation data corrector 310, then combined in a linear combiner 1210 to form Melbourne-Wübbena linear combinations 1220, and finally smoothed over several epochs in a smoother 1410 to form smoothed Melbourne-Wübbena combinations 1420. Alternatively, smoothing can be performed on the original observations or any other linear combination of the original observations before generating the MW linear combination. In any case, the resulting smoothed MW observations are input into the processor 1225 for estimating MW satellite biases 1430 and WL ambiguities 1435 as in the embodiments of Fig. 12 and Fig. 14.

[0095] Smoothing means combining multiple observations over time, for example by a simple averaging operation, to obtain an observation with reduced noise. Mean-value smoothing is performed to reduce the multipath error present in (10), which is not explicitly represented by a model in the processor 1225, as in the model equations or modeling equations (11) and (12). Smoothing is motivated by the expectation that the mean-value observation is nearly constant over short time periods, since the mean-value observation consists only of hardware biases and a (constant) ambiguity expression. A reasonable smoothing interval is, for example, 900 seconds. An additional benefit of smoothing the mean-value observation is that an observational variance for the smoothed observation can be derived from the input data by the variance of the mean. σobs,MW2=∑t=1n(xt−x¯)2(n−1)n with x¯=1n∑t=1nxt where xt The mean observation at smoothing epoch t is given by and n is the number of observations used in the smoothing interval. To ensure that this variance truly reflects a multipath, and not merely an insufficient number of potentially unreliable observations within the smoothing interval, it is advantageous to accept a smoothed observation as filter input only if a minimal number of observations are available, for example, 80% of the theoretical maximum. It should be noted that the statistical data representing the mean and variance of the Melbourne-Wübbena observation must be reset in the case of an uncorrected cycle slip, as this observation contains an ambiguity expression. Naturally, a cycle slip also requires resetting the corresponding ambiguity in the filter.

[0096] If smoothing is performed by a simple averaging operation over a fixed time interval, it implies different data rates in the process. Observations 1205 arrive at a high data rate, while smoothed MW observations 1420 are forwarded to the processor at a low data rate. This type of smoothing has the advantage that observations input to the processor are uncorrelated and can therefore be handled in a mathematically correct manner. The alternative of using some kind of (weighted) moving average allows one to remain at a single (high) data rate, but has the disadvantage that the resulting smoothed observations are mathematically correlated. Part 7.4 MW Process: MW Bias Condition

[0097] Fig. Figure 16 shows a process 1600 with the addition of one or more MW bias conditions to the process of Fig. 15, which can be added similarly to those in Fig. 12, Fig. 14 embodiments shown. At least one MW bias condition 1605 such as (32) is input into the processor to reduce the rank defect in model equations (11) or (12).

[0098] The rank defect in (11) or (12) becomes apparent by counting the number of observations and the number of unknowns in these equations. For example, in (11), if there are i = 1,...,m stations and j = 1,...,n satellites, and it is assumed that all satellites are seen at all stations, there will be m·n Melbourne-Wübbena observations. However, at the same time, there are also m·n unknown ambiguities. Ni,WLj in addition to m receiver biases b i,MW and n satellite biases bMWj, which results in m·n+m+n unknowns. Therefore, the system of equations defined by (11) can only be solved if the number of arbitrary conditions or boundary conditions introduced into the system is the number of unknowns minus the number of observations, i.e., (m·n+m+n)-(m·n)=m+n.

[0099] Most of these conditions should be ambiguity conditions, as the following consideration demonstrates. For n satellites, n-1 independent inter-satellite single differences can be generated. In the same way, m-1 inter-station single differences can be derived from m stations. In a station-satellite double difference (DD), these independent single differences are combined, resulting in (m-1)·(n-1)=m·n-(m+n-1) double-difference observations. Since, as in (13), the DD ambiguities are uniquely determined by the DD-MW observations, the difference between the m·n ambiguities in (11) and the m·n-(m+n-1) unique DD ambiguities should be restricted, resulting in m+n-1 ambiguity conditions. Therefore, of the m+n required conditions, all but one should be ambiguity conditions. The remaining arbitrary condition should be a condition orThe restriction regarding the bias is a limitation. This statement remains true in the more general case where not all satellites are seen at all stations, and therefore the number of required conditions can no longer be counted in the simple manner shown. The condition regarding the bias itself is any equation such as (32) or, more generally, of the form . ∑iaibi,MW+∑jajbMWj=b with ai,aj,b∈R

[0100] In the single-difference case (12), the condition regarding the bias is simpler. The state vector of the satellite bias does not contain all possible SD biases, but only the independent ones. This is achieved by using an arbitrary reference satellite and selecting only the SD bias for the reference as states. To be prepared for a change of reference satellite in the event that the old reference satellite is no longer observed, it is preferable to also have a state for the reference satellite. However, the SD bias of the reference to itself must be zero. bMWjrefjref=bMWjref−bMWjref≡0

[0101] This equation is added as a condition. It should be noted, however, that the SD bias states are not necessarily interpreted as SDs for a physical reference satellite. It is also possible to have an artificial reference satellite with a bias that references the bias of the physical satellites (this ensures that the artificial satellite is linked to the physical satellites). ∑jajbMWj=b with aj,b∈R

[0102] By specifying arbitrary values ​​for a j , b (with at least one a j ≠ 0) and inserting (37) as a condition into (12), the information about the bias of the reference station enters the system.

[0103] With knowledge of MW satellite biases (as derived from the system proposed here) from a different source, it is also reasonable to introduce more than one bias condition into the system. For example, if all MW satellite biases are bounded, in a single-difference approach it is not necessary to introduce any ambiguity conditions into the system, since (12) can be written as Φi,WLj1j2−Pi,NLj1j2+bi,MWj1j2=λWLNi,WLj1j2

[0104] Therefore, all SD ambiguities Ni,WLj1j2 unambiguously determined with knowledge of the SD-MW satellite biases. It is precisely this relationship that helps a rover receiver resolve its WL ambiguities with the help of the MW satellite biases derived here.

[0105] In the undifferentiated approach, an ambiguity condition is introduced per station when the MW satellite biases are introduced for all satellites as conditions into the system.

[0106] All bias conditions for handling the rank defect in model equations (11) or (12) are circumvented if an additional ambiguity condition is introduced instead. However, this additional ambiguity condition is not arbitrary. It is chosen such that the double difference relation (13) is satisfied. However, (13) does not include the multipath not represented as a model and only determines a floating-point value for the DD ambiguity. Therefore, deriving an integer DD ambiguity value from (13) is prone to errors.

[0107] To better distinguish between arbitrary ambiguity conditions and ambiguity conditions that must satisfy the DD ambiguity relation (13), we usually call the second type of conditions ambiguity fixes. While conditions are arbitrary and do not depend on actual observation, fixes do. Conditions cannot be performed for an incorrect value, but fixes can. Which conditions are restricted to an arbitrary value is described in Part 7.6. Part 7.5 MW Bias Process: WL Ambiguity Conditions

[0108] Fig. Figure 17 shows a process 1700 with the addition of one or more WL ambiguity conditions to the process of Fig. 16, which can be similarly added to those in Fig. 12, Fig. 14 and Fig. 15 embodiments shown. At least one WL ambiguity integer condition 1705 as in equation (33) is fed into the processor 1225 to further reduce the rank defect in model equations (11) or (12). As for Fig. Equation 16 represents the correct number of arbitrary or arbitrary ambiguity conditions in a network with i=1,...m stations and j=1,...,n satellites, m+n-1, where all satellites are seen by all stations. However, in a global network with reference stations distributed across the entire Earth, not all satellites are seen by all stations. For this case, selecting the correct number of arbitrary ambiguity conditions and determining the exact combinations that are not restricted by the DD ambiguity relation (13), described in Part 7.6, is necessary.

[0109] Although the restricted ambiguities, which are not restricted by the DD ambiguity relation (13), could be restricted to any value to reduce the rank effect in model equations (11) or (12), it is desirable to restrict these ambiguities to an integer value so that the integer nature of the ambiguities is introduced into the system. This helps later when integer values ​​consistent with (13) are sought for the remaining unrestricted floating ambiguities, and at which these ambiguities can be fixed.

[0110] In a Kalman-filter approach, where equations (11) or (12) are extended by equations for the initial values ​​of these parameters, there is always a well-defined floating solution for all parameters (which, however, exhibits large variance if the initial variances of the parameters were also chosen with large values). In this case, it is reasonable to restrict the ambiguities to the nearest integer of their Kalman-filter floating solution, as this disturbs the filter in the least way and yields the solution that is closest to the initial values ​​of the parameters. It is also advantageous to restrict the ambiguities one at a time, checking the updated floating ambiguity of the next ambiguity to be restricted after each condition.Such a procedure helps to stabilize the filter in cases of network failures, where many ambiguities are lost, receiver biases are represented by a model as white noise parameters, and already converged satellite biases have a defined value. Part 7.6 MW Bias Process: Determining WL Ambiguity Conditions

[0111] Fig. Figure 18 shows a process 1800 with the addition of determining one or more WL ambiguity conditions for the process of Fig. 17, so that under- and over-limiting of the model equations (11) or (12) is avoided.

[0112] Under-restriction means that not enough conditions have been introduced to handle the rank defect in (11) or (12). Over-restriction means that arbitrary conditions are introduced that are inconsistent with the DD ambiguity relation (13). Therefore, a system can be both over- and under-restricted at the same time.

[0113] The MW observation input 1420 defines an observation graph 1805, i.e., a number of edges given by observed station-satellite connections. Based on this graph, a spanning tree (ST) 1815 is generated by an ST generator 1810, which connects all stations and satellites (in the undifferentiated case (11)) or only all satellites (in the inter-satellite single-differentiated case (12)) without introducing any loops. The resulting ST 1815 defines the WL ambiguity conditions 1705, i.e., which WL ambiguities are bounded.

[0114] Fig. Figure 19A shows at 1900 how observed station-satellite connections can be interpreted as an abstract graph, i.e., a number of nodes connected by edges. The stations in the lower part of Fig. 19A and the satellites in the upper part of Fig. 19A are identified as nodes, and the station-satellite pairs (each pair corresponding to observations at a station of a signal from a satellite) as edges. The abstract graph 1910 of 19B does not distinguish further between stations and satellites and instead shows the edges as connections between nodes or vertices.

[0115] In graph theory, a tree is a graph without closed loops. An example is given in 1920. Fig. Figure 19C shows the tree marked with thick lines. A spanning tree on a graph is a tree that includes all nodes, as shown in Fig. 19C, connects (or spans).

[0116] Instead of building the spanning tree based on the current observation graph, it can alternatively be based on all station-satellite ambiguities currently in the filter. These ambiguities correspond to station-satellite connections that were observed in the past but are not necessarily being observed in the current epoch. We call the station-satellite connections defined by the filter ambiguities the filter graph. Note that the length of time ambiguities are retained in the filter after they are no longer being observed is somewhat arbitrary.If a fixed slot management system is used for ambiguities in the filter, which maintains a maximum number of ambiguities for each station, such that a newly observed ambiguity for an ascending satellite will discard the oldest ambiguity if all slots are already in use, then the duration for which a certain ambiguity is held does not need to be specified. It will vary for each satellite at each station. However, such slot management guarantees that after a certain period, each station will hold the same number of ambiguities.

[0117] In general, the filter graph contains more station-satellite links than the observation graph. It additionally includes stations that are no longer observed (which often occurs for short periods in a global network), satellites that are no longer observed for a station (for example, because a satellite became unusable for a short period), or only station-satellite links that have fallen below the altitude interval. Working on the filter graph is of particular importance if the ambiguity setting described later is also performed on the filter graph, and ambiguity conditions and settings are added to the original filter rather than a filter copy.

[0118] In the individually differentiated observation graph from 1960 by Fig. In 19G, two satellites are usually connected by multiple edges, since the two satellites are typically observed at multiple stations. Each edge connecting two satellites corresponds to a (at least historically) observed satellite-station-satellite connection, i.e., a single-difference observation. Naturally, the SD filter graph from 1970 also contains Fig. 19H has more edges than the SD observation graph from 1960.

[0119] Restricting the ambiguities, determined by a spanning tree over the observation or the filter graph, can prevent under- and over-restriction of model equations (11) or (12). This is shown for the undifferentiated case in Fig. 19D. A spanning tree (ST) on the observation graph or filter graph connects all nodes without introducing loops (see highlighted edges in Fig. 19C). Fig. In addition to the spanning tree edges (in dark gray) which are restricted to an arbitrary integer value, 19D also shows a single-satellite bias condition S1, shown in dark gray, at 1930. The satellite bias is visualized as a circle because its contribution to the observation is the same for all receiver observations of that satellite.

[0120] Restricting the ST edges together with an additional satellite bias S1 allows the underdetermined linear system of equations (11) to be solved: the observation R1-S1 together with the satellite bias condition S1 and the ambiguity condition R1-S1 allows receiver bias R1 to be uniquely solved (compare equation (11)). Once receiver bias R1 is known, the observation R1-S2 together with the ambiguity condition R1-S2 allows satellite bias S2 to be solved, or rather, solved for satellite bias S2. In the same way, all other receiver and satellite biases can be computed with the help of the ST-restricted ambiguities. The spanning property of the ST ensures that all satellite and receiver biases are reached, while the tree property ensures that there are no loops that would impose constraints.Once all satellite and receiver biases are known, the remaining ambiguities (for example, R2-S1, R2-S4 and R2-S5) can be resolved. Fig. 19D) can be calculated directly from the remaining observations one after the other.

[0121] In the SD case, shown in Fig. 19G and Fig. In 19H, the reasoning is quite similar. Restricting an SD satellite bias to an arbitrary value (for example, restricting the bias of a physical reference satellite to zero) allows the SD satellite bias of the next satellite to be determined using an SD observation between the first and second satellites and the ambiguity condition of the SD spanning tree between the two satellites (compare equation (12)). Once the second satellite bias is known, the third bias can be calculated. In the same way, all other satellite biases are determined using the SD spanning tree conditions. By adding an ambiguity condition per station to any satellite, all remaining SD ambiguities (differentiated individually against a station-specific reference satellite) in the filter can be resolved successively.

[0122] The relationship underlying this description between an undifferentiated (=zero-differentiated (ZD)) Spanning Tree 1975 and an SD-Spanning Tree 1980 is in Fig. As shown in 19I, connecting each station to a single satellite by introducing one ambiguity condition per satellite and adding to these conditions those given by an ST on the SD observation graph (or filter graph) defines the same conditions given by an ST on a ZD observation graph (or filter graph) in 1985. Constructing a spanning tree on a graph is not a unique process. For a given graph, there exist many trees that spann the graph. To make the generation of a spanning tree more unambiguous, the use of a minimum spanning tree (with reference to some criteria) is proposed in Part 7.7. Part 7.7 MW Bias Process: Minimum Spanning Tree

[0123] Fig. 21A shows a Spanning Tree (ST) on an undifferentiated observation graph at 2110. Fig. 21B shows a minimum spanning tree (MST) (Cormen, Leiserson, Rivest, & Stein, 2001) at 2120 on the undifferentiated observation graph of Fig. 21A. Fig. 20 shows the ST producer 1810 from Fig. 18, which is replaced by an MST generator 2010. To construct an MST on a graph, each edge has an edge weight, resulting in a so-called weighted graph. The MST is defined as the spanning tree with the total minimum edge weight. The edge weight can be assigned in various ways. One possible approach is based on the current receiver-satellite geometry and therefore uses the station positions 1250 and the satellite positions 1245 as inputs. The connections between the coarse receiver positions 1250, the transmission satellite orbits 1245, and the MST generator 2010 are shown in Fig. 20 shown.

[0124] The edges in the undifferentiated observation graph or filter graph are defined by station-satellite links. In some embodiments, the edge weighting is the geometric distance from receiver to satellite or a satellite elevation angle-related quantity (such as the inverse elevation angle or the zenith distance (=90° elevation)).

[0125] The edges in the individually differentiated observation graph or filter graph are defined by satellite-receiver-satellite links connecting two different satellites via a station. In some embodiments, the edge weighting is the geometric distance from satellite to receiver to satellite, or a combination of the elevations at which the two satellites are seen by the receiver. In some embodiments, the combination of the two elevations is the sum of the inverse elevations, the sum of the zenith distances, the minimum inverse elevation, or the minimum zenith distance.

[0126] In Fig. 21A and Fig. In section 21B, the ST and MST edges are marked with an "X". The ST 2110 in Fig. 21A and the MST 2120 in Fig. 21B are identical, and the ST 2130 in Fig. 21C and the MST 2140 in Fig. 21D are identical, which is shown by the fact that every ST can be obtained as an MST by defining suitable edge weights.

[0127] An MST is well-defined (i.e., unique) if all edge weights are distinct. Assigning weights to edges allows control over how the spanning tree is generated. In embodiments using geometrically based weights, the MST is generated in a way that favors the highest satellites (those with the smallest geometric separation and smallest zenith distance, or smallest value of 1 / elevation). These are also the station-satellite links least affected by the multipath (not represented as a model). In these embodiments, the weights favor binding edges that should push the smallest multipath to other links when ambiguities are restricted to integer values.In embodiments that use low-elevation station-satellite links and a high multipath for constraints, the multipath is shifted to links with higher-elevation satellites. This can lead to the non-intuitive effect that ambiguities regarding higher-elevation satellites become more difficult to resolve to an integer value.

[0128] Generating a mean square root (MST) on a given weighted graph is a standard task in computer science, for which very efficient algorithms exist. Examples include the algorithms of Kruskal, Prim, and Boruvka.

[0129] Fig. Figure 22 shows an alternative way of selecting the edge weights of the observation graph or filter graph on which the MST (defining the restricted ambiguities) is generated. In some embodiments, the edge weights are derived from the ambiguity information in the filter, i.e., from the values ​​of the WL ambiguities 1435 or from the variances 2210 of the WL ambiguities, or from both.

[0130] A particularly interesting edge weighting used in some embodiments is the distance of the corresponding ambiguity to its nearest integer value. In this way, the MST selects the ambiguities for bounding that span the observation / filter graph and are as close as possible to an integer. Therefore, the states in the processor are minimally affected by the integer-related conditions. An additional advantage is that the conditions of the previous epoch are also favored in the new epoch, since the distance to an integer is zero. This prevents over-bounding of the filter when the conditions are applied directly to the filter and no filter copy is used.The same goal is achieved by using the variance of the ambiguity as an edge weighting; restricted ambiguities have a variance equal to the size of the restricting variance, for example 10. -30 m 2and are therefore favored in MST generation in some embodiments. In some embodiments, any combination of ambiguity distance to integer and ambiguity variance is used as an edge weight. In some embodiments, a combination of an edge weight derived from the station-satellite geometry and from the ambiguity information is used. For example, in some embodiments, a geometrically motivated weight is selected in the first epoch (to minimize the effects of multipaths not represented by the model and to ensure that conditions remain in the system for a long time), and an ambiguity-derived weight is selected in later epochs (to avoid over-constraints). Part 7.8 MW Bias Process: Determining WL Ambiguity

[0131] Fig. Figure 23 shows a setting of the WL ambiguities before they are sent (for example, for use in the phase-time signal processor 335 or orbit processor 300). In some embodiments, the WL ambiguity state values ​​1435, together with at least the WL ambiguity variances 2210, are passed from the filter to an ambiguity-setting module 2305. The ambiguity-setting module 2305 outputs the fixed WL ambiguities 2310.

[0132] The ambiguity detection module 2305 can be implemented in a variety of ways:

[0133] Threshold-based integer rounding: In some embodiments, a simple setting module checks each individual ambiguity to determine if it is closer to an integer than a given threshold (for example, more than a = 0.12 WL cycles). If the standard deviation of the ambiguity is also below a second given threshold (for example, σ = 0.04 such that α = 3σ), the ambiguity is rounded to the nearest integer for setting. Ambiguities that do not meet these setting criteria remain unset. In some embodiments, the satellite elevation angle corresponding to the ambiguity is considered as an additional setting criterion, so that, for example, only ambiguities greater than 15° are set.

[0134] Optimized sequence, threshold-based integer bootstrapping: A slightly more advanced approach, used in some embodiments, defines ambiguities in a sequential, or successive, manner. After defining one component of the ambiguity vector, the definition is reintroduced into the filter, so that all other undefined ambiguity components are affected by their correlations to the defined ambiguity. Then, the next ambiguity is checked for satisfying the definition criteria of a distance to an integer and standard deviation. In some embodiments, the sequence for checking ambiguity components is selected in an optimal way by ordering the ambiguities with respect to a cost function, for example, distance to an integer plus three times the standard deviation.In this way, the most reliable ambiguities are determined first. After each determination, the cost function for all ambiguities is re-evaluated. Following this, the ambiguity with the lowest cost is checked again to see if it meets the determination criteria. The determination process stops if the best candidate for determining the ambiguity does not meet the determination criteria. All remaining ambiguities remain undefined.

[0135] Integer least squares (generalized) partial fixing: A more sophisticated approach, used in some embodiments, considers the covariance information of the ambiguities from the filter. The best integer candidate N1 is the closest integer vector to the least squares floating ambiguity vector N̂ ∈ R. v in the metric defined by the ambiguity part of the (unbounded) state covariance matrix P N̂ ∈ R v × Rv , both of which are obtained from the filter, i.e. N1=arg minN∈z′(N−N^)′PN^−1(N−N^)

[0136] However, since the observation input to the filter is only precise to a certain extent due to measurement noise, the resulting estimated floating ambiguity vector N is also only reliable to a certain extent. Therefore, a slightly different N̂ can correspond to a different N ∈ Z. v This leads to the minimization of (39). Therefore, in some embodiments, the best integer candidate is replaced with, for example, the second-best integer candidate by feeding other noisy measurements (e.g., from other receivers) into the filter. To identify the reliable components in the ambiguity vector that can be fixed to a unique integer with a high probability, the minimized size is (N−N^)′PN^−1(N−N^) compared in some embodiments among the best integer candidates in a statistical test such as the ratio test. If N i The fact that the i-th best (i > 1) integer candidate is implied that (Ni−N^)′PN^−1(Ni−N^)>(N1−N^)′PN^−1(N1−N^), or Fi:=(Ni−N^)′PN^−1(Ni−N^)(N1−N^)′PN^−1(N1−N^)>1

[0137] The quotient in (40) is a random variable that follows an F-distribution. Some embodiments essentially follow the description in (Press, Teukolsky, Vetterling, & Flannery, 1996). The probability that F i would be as big as it is, if (Ni−N^)′PN^−1(Ni−N^) is smaller than (N1−N^)′PN^−1(N1−N^), is referred to as Q(F i |v,v), whose relationship to the beta function and precise algorithm determination is given in (Press, Teukolsky, Vetterling, & Flannery, 1996). In other words, Q(F i|v,v) the significance level at which the hypothesis (Ni−N^)′PN^−1(Ni−N^)<(Ni−N^)′PN^−1(Ni−N^) can be rejected. Therefore, any candidate for whom, for example, Q(F) i |v,v)≥0.05 is considered to be explained as well as N1. The first candidate i0+1 for which Q{F i0+1 If |v,v) < 0.05 is accepted as significantly worse than N1.

[0138] Then all the components N1, N2, ... 'N i0 Vectors with the same value are considered reliable integer determinations. The components in which these ambiguity vectors differ should not be fixed to an integer. However, certain linear combinations may exist among these components that are the same for all vectors N1,N. 2, ...,N i0 These linear combinations can also be reliably fixed to an integer.

[0139] In some embodiments, the determination of the best integer candidate vectors is carried out using the efficient LAMBDA method (Teunissen, 1995).

[0140] Best integer equivariant approach: In some embodiments, the components of the high-dimensional ambiguity vector are fixed to floating-point values, which are given by a linear combination of the best integer candidates. Fig. 23B shows that the WL ambiguities are transmitted (for example, for use in the phase-time signal processor 335 or orbit processor 300).

[0141] In these embodiments, the WL ambiguity state values ​​1435, together with the ambiguity-variance-covariance matrix 2210, are passed from the filter to an integer ambiguity search module 2320. The integer ambiguity search module 2320 outputs a number of integer ambiguity candidate sets 2323, which are then passed to an ambiguity combiner module 2325. This module also receives the least-squares ambiguity glide solution 1435 and the ambiguity-variance-covariance matrix 2210 from the filter as input. The least-squares ambiguity glide solution 1435, together with the ambiguity-variance-covariance matrix 2210, is used to form a weight for each integer ambiguity candidate. In the ambiguity combiner module 2325, the weighted integer ambiguity candidates are summed.The output is a fixed WL ambiguity vector 2330, which can then be passed on to the phase-time signal processor 335 and orbit processor 330.

[0142] To derive the weights for the integer ambiguity vectors, it should be noted that the least-squares ambiguity glide vector N̂ is the expected value of a multidimensional Gaussian probability function p(N), i.e. N^=∫N∈RuN p(N) dNwithp(N)=e−12(N−N^)′PN^−1(N−N^)∫N∈Rve−12(N−N^)′PN^−1(N−N^)dN

[0143] Therefore, an ambiguity expectation value that recognizes the integer nature of ambiguities is given by N⌣:=∑N∈ZνN p⌣(N)withp¯(N)=e−12(N−N^)′PN^−1(N−N^)∑N∈Zνe−12(N−N^)′PN^−1(N−N^)

[0144] Since the summation over the entire integer lattice N ∈ Z vSince this cannot be calculated in practice, in some embodiments the sum is restricted to the best integer ambiguity candidates. N⌣≈∑i with Fi<1+εNi p⌣(Ni)withp⌣(Ni)≈e−12(Ni−N^)′PN^−1(Ni−N^)∑i with Fi<1+εe−12(Ni−N^)′PN^−1(Ni−N^) with F i of (40). The p̌(N i ) are the desired weights for the best integer ambiguity candidates. A reasonable value for ε follows from the following consideration. In sum (42), ambiguity vectors N can be neglected if the relative weight for the largest p̌(N) is small, i.e., if p̌(N) / p̌(N1) ≤ δ, for example, with δ = e -q ; q=25. In other words, all ambiguity vectors with p̌(N)|p̌(N1)>δ must be identified. Writing out this condition for N=N i with the definition for p̌(N) of (42) leads to δ <p⌣(Ni)p⌣(N1)=e−12(Ni−N^)′PN^−1(Ni−N^)e−12(N1−N^)′PN^−1(N1−N^)=e12(N1−N^)′PN^−1(N1−N^)−12(Ni−N^)′PN^−1(Ni−N^)⇔lnδ=−q<12(N1−N^)tPN^−1(N1−N^)−12(Ni−N^)tPN^−1(Ni−N^)⇔−q12(N1−N^)tPN^−1(N1−N^)<1−12(Ni−N^)tPN^−1(Ni−N^)12(N1−N^)tPN^−1(N1−N^)︸=:Fi⇔Fi<1+q12(N1−N^)tPN^−1(N1−N^) therefore ε:=q12(N1−N^)′PN^−1(N1−N^) a reasonable value for ε.

[0145] The p̌(N i The desired weights for the best integer ambiguity candidates are shown. The ambiguity candidates themselves can be efficiently determined using the Lambda method (Teunissen, 1995). Part 7.9 MW Bias Process: Using Fixed WL Ambiguities

[0146] Fig. Figure 24A shows an embodiment 2400. In this way, the estimated MW biases 1430 are made consistent with the fixed WL ambiguities 2330. The fixed-nature MW satellite biases from the network are transferred to the rover receiver, where they help in establishing WL ambiguities.

[0147] The fixed WL ambiguities 2330 can be introduced into the processor 1225 in several different ways. Fig. 24B, Fig. 24C and Fig. Figure 24D shows details of these possible realizations of the 1225 processor, which differ in the way the fixed WL ambiguities are traced back to the MW bias estimation process.

[0148] In embodiment 2400, the processor 1225 comprises a single sequential filter 1410 (such as a Kalmann filter) that holds states 2415 for satellite MW bias, as well as states 2420 for WL ambiguities and—in the case of the undifferentiated observation model (11)—also states 2425 for receiver WM bias. In the single-differentiated (SD) observation model (12), no receiver bias states occur. In addition to these states, which contain the values ​​of the least-squares best solution of the model parameters for the given observations, the filter also contains variance-covariance (VC) information 2430 of the states. The VC information is usually given as a symmetric matrix and shows the uncertainty in the states and their correlations. It does not depend directly on the observations, but only on the observation variance, process noise, observation model, and Initial variations.However, since the observational variance is derived from the observations, if smoothing is possible (see Part 7.3), there can also be an indirect dependence of the VC matrix 2430 on the observations.

[0149] The filter input includes MW observations (for example, smoothed MW observations 1420) and satellite MW bias process noise 1240 (see part 7.2), an MW bias condition 1605 (see part 7.4), integer WL ambiguity conditions 1705 (see part 7.5) and optional shifts of a bias and an ambiguity shifter 24 (discussed in part 7.10).

[0150] The primary filter output includes the satellite MW biases 1430, the (floating) WL ambiguities 1430, and the (unrestricted) WL ambiguity part 2210 of the VC matrix 2430. The ambiguity information is passed to an ambiguity determination module (see Part 7.8), which outputs (floating or integer) fixed WL ambiguities 2310. These fixed WL ambiguities 2310 are output for use in the orbit processor 330 and the phase-time signal processor 335. Additionally, the fixed WL ambiguities 2310 are reintroduced into the filter by adding them as pseudo-observations with a very small observation variance (of, for example, 10). -30 m 2The resulting satellite MW biases 1430 are output to the rover and optionally to the orbit processor 330 and the phase-time signal processor 335. Since they are consistent with the fixed WL ambiguities, we call them fixed MW biases.

[0151] It should be noted that if an ambiguity was set to an incorrect integer, the incorrect ambiguity would remain in filter 2410 until a cycle shift or cycle jerk occurs, at which point an ambiguity is either detected or removed from the filter (such as when it is no longer observed for a certain period or another ambiguity takes its ambiguity slot in filter 2410). If this were to occur, the MW biases would be disturbed for a long time. However, an advantage of this approach is that ambiguity settings that were initially set to an integer also remain in the filter if the satellites were observed at high elevations but have since moved to lower elevations and can no longer be set, or are no longer observed at all.These ambiguity settings when aligning satellites can greatly stabilize the solution of MW biases.

[0152] It should also be noted that a floating ambiguity should not be defined with a very small variance (of, for example, 10). -30 m 2 ) in the single filter 2410, since in this way new observations can no longer improve the ambiguity states by bringing them closer to an integer. In some embodiments, the floating ambiguity is fixed with a variance that depends on its distance to an integer, such that the observational variance tends to zero when the distance to the nearest integer tends to zero and to infinity when the distance to the nearest integer tends to 0.5. However, the approaches used in embodiments of are for fixing floating ambiguities Fig. 24C and Fig. 24D can be used, which is more suitable.

[0153] In embodiment 2440 of Fig. In the 24C version, the 1225 processor includes two sequential filters, 2445 and 2450, with the process flow for the first filter, 2445, being almost identical to that of the 2410 filter. Fig. 24B. The difference is that no best WL ambiguities 1430 are fed back into filter 2445. Instead, each time new fixed WL ambiguities 2310 are available (for example, after each observation update), a filter copy 2450 of the first filter 2445 is made, and then the fixed WL ambiguities 2310 are introduced as pseudo-observations into filter copy 2450. The original filter 2445 thus remains unaffected, so no false biases can be introduced into it. Filter copy 2450 outputs fixed satellite MW biases 2455 (for example, as MW bias 345).

[0154] One disadvantage of this approach is that only currently observed ambiguities can be determined and introduced into filter copy 2550. All previous ambiguity determinations are lost. However, this is a preferred processing method when the entire ambiguity space is analyzed at once, as is done in the integer least squares partial determination and candidate integer combination approaches (see Part 7.8).

[0155] embodiment 2460 of Fig. Figure 24D shows an alternative approach for feeding the fixed WL ambiguities into the estimation process. Here, the fixed WL ambiguities 2310 are passed to an ambiguity subtraction module 2665, which reduces the MW observations 1420 by the ambiguities. The resulting ambiguity-reduced MW observations 2670 are fed into a second filter 2475, which has no ambiguity states but only satellite MW bias states 2480 and—in the undifferentiated approach (11)—also receiver MW bias states 2485. This second filter 2475 requires only a single MW bias condition 2490 and process noise on satellite MW biases 2480 as additional inputs. If biases are moved or shifted in the first filter 2440, they must also be moved in the second filter 2475.

[0156] The second filter 2475 outputs fixed satellite MW bias 2450.

[0157] It should be noted that in this approach the ambiguities are not fixed, with a very small observational variance (of, for example, 10). -30 m 2 ), but only with the usual observational variance of the mean square (MW) observations. By inserting observations over time with the same fixed ambiguity, the weak ambiguity condition is tightened more and more. All previous ambiguity settings remain to some extent in the filter. An incorrect setting that is detected after some time will be averaged out or smoothed out. Therefore, it is also quite reasonable to input floating ambiguity-reduced WM observations into the filter.

[0158] Since the second filter 2475 has no ambiguity states, which constitute the majority of states in the first filter, the second filter 2475 is very small and can be updated at a very high rate, for example, every second, without encountering any performance problems. Therefore, in some embodiments, the original MW observations are fed into this filter without any prior smoothing. Part 7.10 MW Bias Process: Shifting MW Biases

[0159] Fig. Figure 25A shows an embodiment 2500 in which the process described in Part 7.8 is extended by an external satellite MW bias shift module 2505. The term external means, in contrast to the shift module described in Fig. Figure 25C shows where the shift is applied to the filter. It should be noted that all the ambiguity restriction and fix-related steps, as well as the correction and smoothing steps, are optional.

[0160] The bias shift module 2505 shifts the satellite MW biases 1430 to generate satellite MW biases 2510 within a desired range of at least one WL cycle. This is possible because, as seen from equation (11), a shift in a satellite bias of n WL cycles is absorbed by the WL ambiguities corresponding to that satellite, for example Φi,WLj−Pi,NLj=bi,MW−bMWj+λWLNi,WLj=bi,MW−(bMWj+nλWL)︸=:b˜MWj+λWL(Ni,WLj−n)︸=:N˜i,WLj

[0161] Similar shifts are possible for receiver bias.

[0162] Fig. Figure 25B shows the effect of shifting MW biases, as in equation (46). Each MW combination is in Fig. Figure 25B shows the distance between a receiver (for example, one of receivers 2525, 2530, 2535, or 2540) and a satellite 2545. This distance is represented by the sum of a receiver bias (which is the same for all satellites and is therefore visualized as a circle around the receiver, as shown in Figure 2550), a satellite bias (which is the same for all receivers and is therefore visualized as a circle, as shown in Figure 2555, around the satellite), and an ambiguity (which depends on the receiver-satellite pair and is therefore visualized as a bar, as shown in Figure 2560, for the pair of receiver 2525 and satellite 2545). Decreasing the satellite bias by the wavelength of one WL cycle (as shown by a smaller circle, as shown in Figure 2565) increases all related ambiguities by the wavelength of one WL cycle. The receiver biases are unaffected by this operation.

[0163] One advantage of shifting satellite MW biases to a defined range is that the biases can be encoded with a fixed number of bits for a given resolution. This allows the bandwidth required to transfer satellite MW biases to the rover to be reduced, which in some designs is done via expensive satellite links.

[0164] Although all satellite biases bMWjλWL Since the mapping can be set to a fixed range of [-0.5, +0.5[ for a certain period of time, it is preferable to extend this range further, for example to [-1, +1[, to avoid frequent jumps in the MW satellite biases when they leave the defined range. Due to the oscillatory behavior of MW satellite biases, the satellite biases at the boundary of the defined range, near -0.5 or +0.5, can often leave this range. For example, a bias moving towards -0.5-ε is then mapped to +0.5-ε. The bias then oscillates back to +0.5+ε and is then mapped back to -0.5+ε. In practice, it has been found that with a range of [-1, +1[, bias jumps can be avoided for several months.

[0165] It should be noted that MW bias jumps can also be handled on the rover by comparing the last received MW bias value with the previous one. If the values ​​differ by approximately one cycle, a bias jump is detected, and the correct bias is reconstructed. The situation is complicated by the fact that the WL ambiguities, consistent with shifted satellite MW biases, are used in the phase-time signal processor to determine iono-free (IF) biases, which are also transmitted to the rover. Reconstructing the MW bias on the rover after a jump also requires an adjustment of the IF bias. 12(λWL−λNL).

[0166] Fig. 25C shows an alternative to the External Satellite MW Bias Shift Module 2505 from Fig. 25A. Satellite MW biases 1430 and WL ambiguities 1435 are sent to an internal shift module 2580, which determines shifts for MW biases and WL ambiguities based on equation (46) so that the biases are mapped to the desired range. Then all these shifts are applied to the bias and ambiguity states in the filter. In this way, biases only need to be shifted once, whereas in the approach of Fig. 25A the shifts are repeated each time satellite MW biases are issued.

[0167] However, it should be noted that unshifted and shifted WL ambiguities are not permitted to be used simultaneously in a single filter. This is important, for example, when WL ambiguities are passed to the Orbit Processor 330 to set IF ambiguities. If fixed IF ambiguities are newly introduced into the single original filter (and not a filter copy as in Fig. (24C is used), WL ambiguities from different eras are combined in the filter. It must be ensured that the WL ambiguities from different eras are the same. If this is not the case, the corresponding IF ambiguity is reset. Part 7.11 MW Bias Process: Numerical Examples

[0168] The behavior of daily solutions for MW satellite biases was monitored over a 61-day period in June and July 2008, and the difference between each daily solution and the first day of this period (June 1) was recorded. PRN 16 was chosen as the reference satellite with a bias value of 0. All biases were mapped to the interval [0,1]. Drifts of varying magnitudes in the satellite biases are clearly detectable. All of the larger drifts occur for Block IIA satellites. Individual satellites show drifts of approximately 0.2 WL cycles within a month. These values ​​would motivate a satellite bias update perhaps once per day. However, there is a sudden bias jump of almost 0.2 WL cycles for PRN 24 on June 26. The occurrence of such events demonstrates the importance of real-time estimation and transmission of MW satellite biases.

[0169] In another example, the medium-wave satellite biases for the period from October 2nd to 14th, 2008, were continuously processed in a Kalman filter. PRN 16 was again selected as the reference. The result shows that each satellite exhibits its own daily pattern with a kind of repetition as early as 12 hours (the time it takes a GPS satellite to complete one rotation). The variations in the satellite biases are up to approximately 0.16 WL cycles within 6 hours. The difference between the medium-wave satellite biases and the values ​​they had 24 hours earlier shows that the day-to-day repeatability is usually less than 0.03 WL cycles. However, this day-to-day repeatability does not well reflect the larger intraday variation in the medium-wave satellite biases.

[0170] The filtered satellite WL biases depend on their process noise input. With a noise input variance between 10 -6 and 10 -7With squared WL cycles per hour, the periodic behavior and the sudden bias level change on June 26th are clearly shown. These patterns are not detected with lower noise input. Based on this analysis, a process noise input variance of 5 × 10⁻⁶ is estimated. -7 Squared WL cycles per hour are recommended on satellite WL biases. Part 7.12 MW-Bias Process: References

[0171] References relating to the MW bias process include the following: Bierman, G. J. (1977). Factorization Methods for Discrete Sequential Estimation. New York: Academic Press, Inc. Collins, P. (2008). Isolating and Estimating Undifferenced GPS Integer Ambiguities. Proceedings of ION-NTM-2008, (pages 720-732). San Diego, Calif. Collins, P., Gao, Y., Lahaye, F., Héroux, P., MacLeod, K., & Chen, K. (2005). Accessing and Processing Real-Time GPS Corrections for Precise Point Positioning - Some User Considerations. Proceedings of ION-GNSS-2005, (Seiten 1483-1491). Long Beach, Calif. Collins, P., Lahaye, F., Héroux, P., & Bisnath, S. (2008). Precise Point Positioning with Ambiguity Resolution using the Decoupled Clock Model. Proceedings of ION-GNSS-2008. Savannah, Georgia. Cormen, T. H., Leiserson, C. E., Rivest, R. L., & Stein, C. (2001). Kapitel 23: Minimum Spanning Trees. In Introduction to Algorithms (Zweite Auflage, Seiten 561-579). MIT Press und McGraw-Hill. Datta-Barua, S., Walter, T., Blanch, J., & Enge, P. (2007). Bounding Higher Order Ionosphere Errors for the Dual Frequency GPS User. Radio Sci., 43, RS5010, doi:10.1029 / 2007RS003772 . Ge, M., Gendt, G., Rothacher, M., Shi, C., & Liu, J. (2008). Resolution of GPS carrier-phase ambiguities in Precise Point Positioning (PPP) with daily observations. Journal of Geodesy, Band 82, Seiten 389-399. Grewal, M. S., & Andrews, A. P. (2001). Kalman Filtering: Theory and Practice Using MATLAB. New York: Wiley-Interscience. Héroux, P., & Kouba, J. (2001). GPS Precise Point Positioning Using IGS Orbit Products. Phys. Chem. Earth (A), Band 26 (Nr. 6-8), Seiten 572-578. Laurichesse, D., & Mercier, F. (2007). Integer ambiguity resolution on undifferenced GPS phase measurements and its application to PPP. Proceedings of ION-GNSS-2007, (Seiten 839-448). Fort Worth, Texas. Laurichesse, D., Mercier, F., Berthias, J., & Bijac, J. (2008). Real Time Zero-difference Ambiguities Fixing and Absolute RTK. Proceedings of ION-NTM-2008, (Seiten 747-755). San Diego, Calif. Melbourne, W. (1985). The case for ranging in GPS-based geodetic systems. Proceedings of the First International Symposium on Precise Positioning with the Global Positioning System. Band 1, Seiten 373-386. Rockville, Maryland: US Dept. of Commerce. Mervart, L., Lukes, Z., Rocken, C., & Iwabuchi, T. (2008). Precise Point Positioning With Ambiguity Resolution In Real-Time. Proceedings of ION-GNSS-2008. Savannah, Georgia. Morton, Y., van Graas, F., Zhou, Q., & Herdtner, J. (2008). Assessment of the Higher Order Ionosphere Error on Position Solutions. ION GNSS 21st International Meeting of the Satellite Division. Savannah, GA, USA. Press, W. H., Teukolsky, S. A., Vetterling, W. T., & Flannery, B. P. (1996). F-Distribution Probability Function. In Numerical Recipes in C (Seite 229). Cambridge University Press. Schaer, S. (2000, May 09). IGSMAIL-2827: Monitoring (PI-CI) code biases. Heruntergeladen von http: / / igscb.jpl.nasa.gov / mail / igsmail / 2000 / msg00166.html. Teunissen, P. (1995). The least-squares ambiguity decorrelation adjustment: a method for fast GPS integer ambiguity estimation. Journal of Geodesy, Band 70, Nr. 1-2, Seiten 65-82. Wübbena, G. (1985). Software Developments for Geodetic Positioning with GPS using TI 4100 code and carrier measurements. Proceedings of the First International Symposium on Precise Positioning with the Global. Positioning System. Band 1, Seiten 403-412. Rockville, Maryland: US Dept. of Commerce. Zumberge, J., Heflin, M., Jefferson, D., Watkins, M., & Webb, F. (1997). Precise point positioning for the efficient and robust analysis of GPS data from large networks. Journal of Geophysical Research , Band 102 (Nr. B3), Seiten 5005-5018. Teil 8: Orbit Prozessor

[0172] Precise (centimeter-accurate) orbits allow for precise satellite time signal estimation and precise positioning. Orbital accuracy directly influences the final position accuracy of the rover, which uses precise satellite orbits and time signal data (or clock data) as described herein. Part 8.1 Option 1: Use of IGS-published ultrafast orbits or orbits

[0173] The International Geophysical System Service (IGS) provides predicted satellite orbits that can be downloaded from the internet. A description of these orbits can be found in J. Kouba, A Guide to Using International GPS Service (IGS) Products, Geodetic Survey Division, Natural Resources Canada, February 2003, 31 pages, and at http: / / igscb.jpl.nasa.gov / components / prods.html.

[0174] The IGS ultrafast (predicted) orbits, also called IGU orbits, are generated and published four times a day at the 3rd, 9th, 15th, and 21st hours of the UTC day. IGS claims a 5 cm orbit standard deviation, although our analyses have shown that individual satellite orbit errors can be as high as 60 cm. In one case, we observed a 2 m error.

[0175] Market demands for commercial positioning services require precise orbits with errors of less than 3 cm and high reliability and availability. Currently available IGU orbits do not meet these requirements. Nevertheless, they are useful either for positioning applications where the requirements are less stringent or as a cross-check for detecting gross errors in orbits estimated as below. Part 8.2 Option 2: Real-time determination of satellite orbits

[0176] With reference to Fig. 1. Observational data are streamed in real time from globally distributed GNSS reference stations, such as reference stations 105, 110, and 115, to a network processor 140. In some embodiments, the network processor estimates the satellite orbits in real time using a Kalman filter approach in a numerically stable UD filter implementation, as described in G. Bierman, Factorization Methods for Discrete Sequential Estimation, Academic Press, Inc., New York, 1977. Hereinafter, the term "real time" refers to processing immediately after observational data are available, typically in less than one second. The time-dependent filter state x(t) is set as follows. x cr (t) Receiver time signal error x cs ,(t) Satellite time signal error x qs (t) satellite-dependent orbit parameters x(t) x r (t) Receiver positions x2DT (t) zenith-tropospheric delay for each station x EOP (t) Earth orientation parameter (EOP: x p , y p , UTI-UTC or length of day) x AMB Carrier-phase ambiguities x if-bias iono-free biases, whereby x cr (1) The vector includes all receiver time signal errors in the network. Each station has at least one time signal offset, but may also exhibit drift and drift rates, depending on the nature and stability of the station clock or station time signal. The receiver time signals are formed by a model, for example, as white noise processes or as stochastic processes (e.g., random walk, Gauss-Markov), depending on the nature and stability of the station time signal. x cs(t) The vector contains the satellite time signals. The vector includes a clock offset or time signal offset and a drift, but drift and drift rates can vary depending on the type and stability of the satellite time signal or satellite clock (rubidium, cesium, or hydrogen maser). The satellite time signals are represented by a model, for example, as white noise processes or as stochastic processes, depending on the type and stability of the satellite time signal. x qs (t) The vector contains the satellite-dependent dynamic orbit parameters. This includes the satellite positions and velocities and additional force model parameters, which are xqs(t)=[xs(t)x˙s(t)xslr(t)xharm(t)] where x s (t) the satellite position vector (one per satellite) is in the inertial reference frame (X,Y,Z). ẋ s,(t) the satellite velocity vector (one per satellite) is in the inertial reference frame (X,Y,Z). x slr (t) The vector is defined by the solar radiation pressure parameters. It consists of a component in the sun-satellite direction, a second component in the direction of the solar radiation panel axis, and a third component perpendicular to the first two axes. All three components are represented by a model, for example, as stochastic processes. x harm (t) The vector is defined with harmonic coefficients for the Orbit components along a path, radially and transversely to a path, or in a satellite-body fixed coordinate system. They are represented by a model, for example, as stochastic processes. x r (t) is the station position vector in the Earth-centered / Earth-fixed reference frame. Stations can either be fixed or unknown in the estimation process. x ZTD(t) is the vector containing the tropospheric zenith delays, estimated for each station. Tropospheric gradients are also optionally estimated. These parameters are represented, for example, by a model as statistical processes. x EOP (t) are Earth orientation parameters (EPOs) that are routinely estimated in real time. The vector consists of the offsets to the conventional pole (x). p , y p ) and the difference between UT1 and UTC time scales (UT1-UTC or length of day). Precisely known EOP parameters are used for the transition between the inertial and the Earth-fixed reference frames. All three parameters are estimated, for example, as stochastic processes. x AMB Each satellite station link has a unique carrier-phase ambiguity in the filter state. These parameters are represented by a model, for example, as constants. X if-biasIonosphere-free code-carrier bias, one bias per receiver-satellite pair. Code and carrier exhibit biases that differ from receiver to receiver and satellite to satellite, and can vary over time. These parameters are represented, for example, by a model of stochastic processes.

[0177] The ionospheric-free dual-frequency combinations of code and carrier observations have different biases that vary over time. While these parameters can be estimated as additional unknowns in the orbit processor Kalman filter, they are optionally estimated in a separate processor (for example, in a standard time-signal processor 320 as ionospheric-free code-carrier bias 372, shown in Fig. 3) and applied to the pseudo-distance observations used in the orbit processor.

[0178] For

[00172] For linearization purposes, some embodiments have the filter device for estimating differences from a reference trajectory and from initial force model parameters. In these embodiments, the state vector for each satellite is xqs(tk)=[r(tk)−r0(tk)p(tk)−p0(tk)y−y0]=[Δr(tk)Δp(tk)Δy] where x qs (t k ) the satellite state vector at time t k is r(t k ) the satellite position and velocity in the inertial reference frame is r0(t k ) represents the reference trajectory, generated by a numerical orbit integrator p(t k ) the vector with stochastic force model parameters is p0(t k ) the vector with approximate initial stochastic force model parameters is y is the vector with constant force model parameters. y0 is the vector with approximate constant force model parameters and where r(tk)=[xx(tk)x˙x(tk)]

[0179] The prediction in the filter model for the satellite-dependent part is performed using the following relationship [Δr(tk+1)Δp(tk+1)Δy]=[Φrr(tk+1,tk)Φrp(tk+1,tk)Φry(tk+1,tk)0Mk0001][Δr(tk)Δp(tk)Δy]+[0wk0] with Φrr(tk+1,tk)=[∂r(tk+1)∂r(tk)],r(t)=r0(t) Φrp(tk+1,tk)=[∂r(tk+1)∂p(tk)],r(t)=r0(t),p(t)=p0(t) Φry(tk+1,tk)=[∂r(tk+1)∂y],r(t)=r0(t),y=y0

[0180] These matrices are calculated, for example, by integrating the variational equations, as described in the section below regarding numerical orbit integration. M k is the matrix that describes stochastic noise modeling w k is the noise input Part 8.3 Numerical Orbit Integration

[0181] The motion of satellites in orbit can be described by a second-order system of differential equations. x¨=x¨(x,x˙,q) with ẍ Acceleration in the inertial reference frame x Position in the inertial reference frame ẋ Velocity in the inertial reference frame q vector of satellite-dependent force model unknowns and initial position / velocity

[0182] The vector q is defined as q=[r(t0)a] where r(t0) is the inertial position and velocity in the inertial reference frame. a the vector with dynamic force model parameters is

[0183] Satellite position x(t) and velocity ẋ(t) at time t are derived from satellite positions x(t0) and velocity ẋ(t0) at time t0, for example, using numerical integration (a higher-order prediction-corrector method). x˙(t)=x˙(t0)+∫x¨(t)dt x(t)=x(t0)+∫x˙(t)dt

[0184] The real-time filter uses the partial derivatives of the accelerations ẍ(t) with respect to the unknown parameters q. [∂x¨∂q]=[∂x¨∂x][∂x∂q]

[0185] The equation in this example ignores derivatives relating to satellite velocity, since there are no relevant velocity-dependent forces acting on GNSS satellites in mid-Earth orbit.

[0186] The following matrix is ​​calculated for epoch t. i : Φrq(ti,t0)=[∂x∂q(ti,t0)∂x˙∂q(ti,t0)]

[0187] The Matrix for the Next Era i+l. can then be calculated using the chain rule Φrq(ti+1,t0)=Φrq(ti+1,ti)Φrq(ti,t0)

[0188] The partial derivatives of the observations with respect to the unknowns q can be derived again using the chain rule. ∂l∂q=∂l∂rΦrq

[0189] These partials are used in the observation update of the filter.

[0190] The following models are used to calculate the accelerations ẍ acting on the satellite; some embodiments use these for the integration process: 1. The Earth's gravitational field is represented by a model such as the EIGEN-CG01C or the EGM96 model. These are spherical harmonic expansions with very high resolution. Some versions use up to degree and order 12 for orbital integration. 2. Gravitational forces due to the attraction of the sun, moon, and planets. 3. The gravitational forces of the sun and moon, acting on the Earth's shape, will deform it. This effect also alters the gravitational field; it is called the "solid earth tide" effect. Some implementations follow the recommendations of the 2003 IERS Conventions. 4. Some embodiments take into account the Solid Earth Pole Tide, which is caused by the centrifugal effect of polar motion. This tide should not be confused with Solid Earth Tides. Some embodiments follow the recommendations of the IERS Convention 2003. 5. The gravitational forces of the sun and moon acting on the oceans alter the gravitational field; this is called the "ocean tide" effect. Some designs use the CSR3.0 model, recommended by the IERS Conventions 2003. 6. Some embodiments also consider relativistic acceleration, which depends on the position and speed of the satellite. 7. The solar radiation pressure, the acceleration acting on GNSS satellites, is the most difficult to model. Some models consider three components: a first component in the sun-satellite direction, a second component in the direction of the solar radiation panel axis (y-bias), and a third component perpendicular to the first two axes. 8. Some embodiments also represent residual errors as a model, mainly added by insufficient knowledge of the force models over harmonic functions, as described below.

[0191] GNSS satellites such as GPS, GLONASS, GALILEO, and COMPASS are in mean Earth orbit (MEO) at approximately 26,000 km. The following table shows the accelerations acting on GNSS satellites and their effects after one-day orbit integration. Störung Acceleration [m / s²] 2 ] Orbit-Vorhersage-Fehler nach einem Tag [m] Gravitation der Erde 0,59 330000000 Ablattung der Erde 5·10 -5 24000 Lunardirektgezeiten 5·10 -6 2000 Solardirektgezeiten 2·10 -6 900 Higher-degree expressions in geopotential 3·10 -7 300 Direct solar radiation pressure 9·10 -8 100 Radiation pressure along the solar panel axis 5·10 -10 6 Earth's albedo (solar radiation due to reflection by the Earth) 3·10 -9 2 Solid Earth Tides 1·10 -9 0,3 Ocean tides 1·10 -10 0,06 General Relativity 3·10 -10 0,3 Venus in low conjunction 2·10 -10 0,08 Part 8.4 Harmonic Force Modeling

[0192] Some embodiments handle residual errors by introducing a harmonic model with cosine and sine expressions in each of the directions along the orbit, radially to the orbit, or transversely to the orbit. Since the residual values ​​have a period of approximately one revolution of the satellite around the Earth, the harmonics depend on the argument of the satellite's latitude. x¯¨residual=(A1cosu+A2sinuA3cosu+A4sinuA5cosu+A6sinu)=(A1A2A3A4A5A6)(cosusinu) with A1, A2 amplitudes to be estimated for along the railway line A3, A4 amplitudes, which are to be estimated for perpendicular to the track A5, A6 amplitudes to be estimated for a radial component u Argument of the satellite's latitude

[0193] As an alternative to using the orbital, perpendicular, and radial components, the satellite-related system (sun-satellite direction, solar-panel direction, and the normal to both) can be used to model solar radiation pressure. Part 8.5 Transformation from inertial to earth-fixed reference frame

[0194] Some embodiments transform from the inertial to the earth-fixed reference frame using the IAU 2000A precision / nutation formulas, which follow the IERS Conventions 2003. Part 8.6 Earth orientation parameters

[0195] Some implementations take Earth Orientation Parameters (EOPs) from the IERS GPSRAPID.DAILY quick files. These files are provided daily by IERS and are based on a combination of the latest available observed and modeled data (including VLBI 24-hour and intensive, GPS, and AAM (Atmospheric Angular Momentum)). The combination process involves applying systematic corrections and light smoothing to remove high-frequency noise. GPSRAPID.DAILY contains the values ​​for the past 90 days from Bulletin A for x / y pole, UT1-UTC, dPsi, dEps, and their errors, as well as predictions for the next 15 days at daily intervals.

[0196] After interpolation, UT1 is corrected for daily and / or semi-diurnal variations due to ocean tides. The polar motion angles are corrected for daily variations due to tidal gravity for a non-fixed Earth, as well as for daily and / or semi-diurnal variations due to ocean tides. Corrections are calculated according to... ΔA=∑ƒBAƒsinθƒ+CAfcosθƒ,A∈{UT1,xP,yP}

[0197] Amplitudes B Af , C Af The 41 daily and 30 semi-diurnal ocean tidal elements are listed in Tables 8.2 and 8.3 of the IERS Conventions (2003). The EOPs from GPSRAPID.DAILY are introduced as approximations for the real-time estimation process, minimizing linearization errors. Part 8.7 Starting the Real-Time System

[0198] When the system is started for the first time, or after interruptions of more than one day, some embodiments derive initial values ​​for satellite position velocity and force model parameters using satellite transmission ephemerides or IGU orbits. Some embodiments use a least-squares fit to match a numerically integrated orbit to the transmission elements from the satellite navigation message or to the IGU orbit for the satellite. The fitted initial orbit state is then integrated for a time period of up to two days in the future. The satellite positions, velocities, and partials are stored in a "partials" file for use by the real-time system.

[0199] Fig.Figure 26A shows an embodiment of the startup process 2600 for the real-time orbit processor. An orbit integrator 2605 predicts the orbital states (position and velocity) of each satellite in the future, starting from an approximate orbit vector 2615, an initial state taken, for example, from the transmit satellite ephemerides or the IGS ultrafast orbit (IGU) by using numerical integration and modeling of all relevant forces acting on the satellites, such as predicted Earth orientation parameters from IERS. During this process, an orbit prediction 2620 is generated, which contains all predicted orbital states and partial derivatives. In a subsequent step, a fitting process adapts the predicted orbit to the transmit orbit or IGU orbit using a least-squares batch process.This process is iterated until the initial orbital states for the satellites are no longer changed. Then the partials file 2620 is forwarded to the real-time orbit process 2630. Fig. 26B.

[0200] Fig.Figure 26B shows an embodiment of a real-time orbit process 2630. Process 2630 receives values ​​for an initial start vector 2635 from partials file 2635; these values ​​are used to construct predicted observations for an iterative filter 2640 on a regular basis, for example, every 150 seconds. The iterative filter 2640 is, for example, a classical Kalman filter, a UD-factored filter, or a square root information filter (SRIF). The orbit start vector 2635 contains the partial derivatives of the current orbit state and force model parameters with respect to the start vector and EOPs and is used to map the partials of the observations onto the start vector and force model unknowns.The Kalman filter 2640 receives iono-free linear combinations 2645 of observational data from the reference stations in real time, for example, at each epoch, such as once per second, and predicted Earth orientation parameters (EOP) 2610. With each observational update, the Kalman filter 2640 provides improved values ​​for the state vector parameters: receiver time signals 2655 (one per station), zenith-tropospheric delays 2660 (one per station), satellite time signals 2665 (one per satellite), optional satellite time signal rates 2670 (one per satellite), and orbit start vectors 2680 (one orbit start vector per satellite). In the embodiment of . Fig. 26B is supplied with an orbit state vector 2680 from time to time (for example, every epoch, such as once per second) and mapped via the orbit image 2682 onto the present-epoch orbit position / velocity vector 350.

[0201] The orbit image 2682 uses the partial derivatives from the partials file and the following relationship to derive satellite position and velocity to the current epoch t. i from the state vector at epoch t0 ( Fig. 26C). r(ti)=r0(ti)+Φrq(ti,t0)xqs(t0)=r0(ti)+[∂r(ti)∂r(t0)∂r(ti)∂p(t0)∂r(ti)∂y][Δr(t0)Δp(t0)Δy] with r0(t i ) Reference trajectory (position and velocity) generated by a numerical orbit integrator Φ rq (t i ,t0) Partial derivatives of position and velocity at t i with reference to the starting vector x qs (t0) at t i Estimated difference of the starting vector (state vector) at t0

[0202] Next, r(t) will be calculated. i ), which is given in the inertial reference frame, is transformed into the earth-centered / earth-fixed reference frame.

[0203] The present-epoch orbit position / velocity vector 350 in the Earth-centered / Earth-fixed reference frame is forwarded to the standard time signal processor 320, the phase time signal processor 335, and the scheduler 355 of Fig. 3.

[0204] Some embodiments of process 2630 avoid linearization errors by restarting the numerical integration in an orbit integrator 2685 from time to time, such as every 6, 12, or 24 hours. The orbit image 2684 uses the partial derivatives from the partials file and the following relationship to derive a new orbit state vector 2690 in the inertial reference frame at time t0+x hours from the start vector at epoch t0 ( Fig. 26D). r(ti)=r0(ti)+Φrq(ti,t0)xqs(t0)=r0(ti)+[∂r(ti)∂r(t0)∂r(ti)∂p(t0)∂r(ti)∂y][Δr(t0)Δp(t0)Δy] with r0(t i) Reference trajectory (position and velocity) generated by a numerical orbit integrator Φ rq (t i ,t0) Partial position and velocity at t i with reference to the starting vector x qs (t0) at t i Estimated difference of the starting vector (state vector) at t0

[0205] Predicted EOPs 2610 (for example from IERS) and estimated EOPs together with updated new orbit start vector 2690 are used as inputs to the orbit integrator 2685 to transform coordinates between the inertial frame and the Earth-fixed frame, for example following IERS conventions.

[0206] This numerical integration runs in the background and generates a new partials file 2635, which is used to update the predicted observations of the Kalman filter 2640. The starting vector for the numerical integration of the orbit integrator 2675, for example, is the last best estimate of the real-time system, the orbit starting vector 2690. Part 8.8 Defining Ambiguities in a Real-Time Orbit Determination

[0207] The Kalman filter 2640 uses the ionospheric-free dual-frequency combinations 2645 from the L1 and L2 GNSS observations. The ionospheric-free combination typically results in a very short wavelength. λIFNIF:=ƒ12λ1N1−ƒ22λ2N2ƒ12−ƒ22=λ1F1F12−F22︸=:λIF(F1N1−F2N2)︸:=NIFwithƒ1=:F1gcd(ƒ1,ƒ2)ƒ2=:F2gcd(ƒ1,ƒ2)

[0208] The factors F1 and F2 are given in Table 3. For example, for GPS L1 and L2 frequencies with F1=77, F2=60 and λ1=0.1903m, the resulting ion-free wavelength is... λIF=λ1F1F12−F22≈6mm This is below the noise level of the phase observations, so there is no way to directly fix the iono-free ambiguity to the correct integer value. It should be noted that for the iono-free combination of L2 and L5, the iono-free wavelength λ IF ≈12.47cm is large enough for reliable ambiguity resolution.

[0209] To utilize the integer nature of the L1 and L2 ambiguities, one could attempt to resolve for L1 and L2 ambiguities, which is difficult due to the unknown ionospheric contribution to the L1 and L2 carrier phase observations at the stations. A preferred approach is to resolve for carrier phase ambiguities after determining the Widelane ambiguities N.WL := N1 - N2 in a first step. Substituting N2 in (64) with N2 = N1 - N WL results λIFNIF=λNLN1+12(λWL−λNL)NWL with λWL:=cƒ1−ƒ2 and λNL:=cƒ1+ƒ2. Therefore, as soon as N WL is known, (65) can be solved for N1 using the floating-point value of the filter for λ. IF N IF After setting the N1 ambiguity vector to an integer or a weighted average of integer values, this can be reinserted into (65) to obtain a fixed value for λ. IF N IF .

[0210] In some embodiments, the Widelane ambiguities N WLThe ambiguities are calculated in the Widelane bias processor 325 and transferred to the orbit processor 2630, for example, for each epoch. For embodiments in which the orbit processor 2630 processes ionospheric-free observations, the estimated ambiguities are ionospheric-free ambiguities λ. IF N IF .

[0211] For embodiments in which the Widelane ambiguities N WL Equation (65), provided by the MW bias processor 325, can also be used to reformulate the Kalman filter equations (by subtraction). 12(λWL−λNL)NWL from the observations) and direct estimation of the N1 ambiguities.

[0212] As an alternative to resolving the N1 ambiguity with a given integer widelane ambiguity, the above equations can be formulated such that N2 or narrowlane ambiguities are estimated instead; the approaches are equivalent because all these ambiguities are linearly related to each other.

[0213] Some embodiments of the ambiguity “fixing” mechanism 2695 use any suitable techniques known in the prior art, such as simple rounding, bootstrapping, integer least squares based on the lambda method, or best integer equivariant (Verhagen, 2005, Teunissen and Verhagen, 2007).

[0214] The term "setting," as used herein, is intended to include not only setting ambiguities to integer values ​​using techniques such as rounding, bootstrapping, and lambda search, but also calculating a weighted average of integer candidates to preserve the integer nature of the ambiguities if they are not set to integer values. The weighted average approach is described in detail in international patent applications PCT / US2009 / 004476, PCT / US2009 / 004472, PCT / US2009 / 004474, PCT / US2009 / 004473, and PCT / US2009 / 004471, which are incorporated herein by reference.

[0215] Analyses have shown that orbit quality is significantly improved by "setting" the ambiguities, either as integer values ​​or as weighted averages of integer candidates. Part 8.9 Orbit processing at IGS analysis centers

[0216] GNSS orbit determination is performed by a variety of IGS analysis centers. To our knowledge, none of these centers provides a real-time orbit estimation as proposed here.

[0217] The CODE Analysis Center (AC) uses a batch least squares approach, as described in Dach et al. (2007), rather than a sequential filter implementation as described herein. Details regarding the model building are also described in Beutler et al. (1994). The CODE AC participates in the generation of the IGS ultrafast orbit (IGU). Orbit positions correspond to the estimates for the last 24 hours of a 3-day arc analysis plus predictions for the following 24 hours. This is performed every 6 hours, with the processing and prediction time staggered by 6 hours at each step.

[0218] The German Research Centre for Geosciences (GFZ) estimates orbits and calculates a contribution to the IGS ultrafast orbits. The approach is documented in Ge et al. (2005, 2006). Its processing, along with that of the CODE process, is based on a batch least squares approach. GFZ performs ambiguity resolution, but only in a post-processing mode. No attempt has been documented regarding real-time operations involving a sequential filter.

[0219] The European Space Operations Centre (ESOC) of the European Space Agency (ESA) also contributes to the IGS ultrafast orbit calculation. This approach is documented by Romero et al. (2001) and Dow et al. (1999). However, this approach is based solely on predicting satellite orbits; no true real-time processing is performed.

[0220] The Jet Propulsion Laboratory (JPL), USA, determines GPS satellite orbits with its system based on the GIPSY-OASIS II software package, developed for the US Global Positioning System, general satellite tracking, orbit determination, and trajectory studies. Details are published in US Patent 5,963,167 by Lichten et al. The JPL system allows for fully automated operation and delivery of validated daily solutions for orbits, clocks / time signals, station locations, and other information without human intervention. This data contributes to the orbits published by IGS, including the ultrafast orbit service that provides IGU orbits. US Patent 5,828,336 by Yunck et al. describes the implementation of a real-time sequential filter.

[0221] The approach of the embodiments of the present invention differs from that of US Patent 5,828,336 in at least these ways: (1) The approach described in US Patent 5,828,336 appears to use only smoothed pseudo-distances; an exemplary observation update rate given is 5 minutes, (2) the JPL system described in US Patent 5,828,336 appears not to establish any carrier-phase ambiguities, and (3) the approach of US Patent 5,828,336 uses an ionospheric model. Part 8.10 References

[0222] Some references related to orbit processing include the following: Beutler, G., E. Brockmann, W. Gurtner, U. Hugentobler, L. Mervart, and M. Rothacher (1994), Extended Orbit Modeling Techniques at the CODE Processing Center of the International GPS Service for Geodynamics (IGS): Theory and Initial Results, Manuscripta Geodaetica, 19, 367-386, April 1994. Bierman, Gerald J. (1977): Factorization Methods for Discrete Sequential Estimation, Academic Press, Inc., New York Dach, R., U. Hugentobler, P. Fridez, M. Meindl (Editoren) (2007), Documentation of the Bernese GPS Software Version 5.0, Januar 2007 Dow, J., Martin-Mur, T.J., ESA / ESOC Processing Strategy Summary, IGS Central Bureau web site, igscb.jpl.nasa.gov / igscb / center / analysis / esa.acn, 1999 Ge, M., G. Gendt, G.Dick und F.P. Zhang (2005), Improving carrier-phase ambiguity resolution in global GPS, Journal of Geodesy, Band 80, Nummer 4, Juli 2006, DOI: 10.1007 / s00190-005-0447-0 Ge, M., G. Gendt, G.Dick, F.P. Zhang und M. Rothacher (2006), A new data processing strategy for huge GNSS global networks, Journal of Geodesy, Band 79, Nummer 1-3, Juni 2005, DOI: 10.1007 / s00190-006-0044-x Landau, Herbert (1988): On the use of the Global Positioning System in geodesy and geodynamics: modeling, software development and analysis, Issue 36 of the publication series of the surveying program at the University of the Federal Armed Forces Munich, December 1988 Lichten, S., Yoaz Bar-Sever, Winy Bertiger, Michael Heflin, Kenneth Hurst, Ronald J. Muellerschoen, Sien-Chong Wu, Thomas Yunck, James Zumberge (1995) GIPSY-OASIS II: A HIGH PRECISION GPS DATA PROCESSING SYSTEM AND GENERAL SATELLITE ORBIT ANALYSIS TOOL, Proceedings of NASA Technology Transfer Conference, Oct 24-26, 1995, Chicago, Ill. Lichten, Stephen M., Wu Sien-Chong, Hurst Kenneth, Blewitt Geoff, Yunck Thomas, Bar-Sever Yoaz, Zumberge James, Bertiger William I., Muellerschoen Ronald J., Thornton Catherine, Heflin Michael (1999), Analyzing system for global positioning system and general satellite tracking, United States Patent 5963167, October 5, 1999. McCarthy, Dennis D. und Gérard Petit (2003), IERS CONVENTIONS, IERS (International Earth Rotation Service) Technical Note 32, Oktober 2003 Romero, I., C. Garcia Martinez, J.M. Dow, R. Zandbergen (2001), Moving GPS Precise Orbit Determination Towards Real-Time, Proceedings GNSS 2001, Sevilla, Spanien, Mai 2001. Yunck, Thomas P., William I. Bertiger, Stephen M. Lichten, Anthony J. Mannucci, Ronald J. Muellerschoen, Sien-Chong Wu, (1998), Robust real-time wide-area differential GPS navigation, US Patent 5828336, Oktober 27, 1998. Teunissen, P.J.G, S. Verhagen (2009); GNSS Carrier Phase Ambiguity Resolution: Challenges and Open Problems, In M.G. Sideris (ed.); Observing our Changing Earth, International Association of Geodesy Symposia 133, Spinger Verlag Berlin-Heidelberg 2009. Verhagen, Sandra (2005): The GNSS integer ambiguities: estimation and validation, Publications on Geodesy 58, Delft, 2005. 194 pages., ISBN-13: 978 90 6132 290 0. ISBN-10: 90 6132 290 1. Part 9: Phase-matched time signal processor

[0223] With reference to Fig. 3 The phase-time signal processor or phase clock processor 335 receives an input MW bias. bMWj 345 (one per satellite) and / or Widelane ambiguities Ni,WLj 340 (one per satellite receiver pair), precise orbit information 350 (one current orbit position / velocity per satellite), tropospheric information 370 (one tropospheric zenith delay per station), code-matched low-rate time signals 365 (one code-matched low-rate time signal error per satellite), and reference station GNSS observation data 305 or 315. The phase-time signal processor 335 generates from these inputs the computed code-matched high-rate time signals 375 (one code-matched high-rate time signal error per satellite) and the phase-matched high-rate time signals 370 (one phase-matched high-rate time signal error per satellite) and passes on the MW bias 345 (one per satellite). Part 9.1 Determining WL ambiguities of MW biases

[0224] Neglecting one multipath, a useful property of the Melbourne-Wübbena (MW) linear combination is Φi,WLj−Pi,NLj, that besides some remaining noise εi,MWj, only the satellite MW biases bMWj and the receiver MW bias b i,MW and the Widelane ambiguity expression λWLNi,WLj What remains: Φi,WLj−Pi,NLj=bi,MW−bMWj+λWLNi,WLj+εi,MWj

[0225] To reduce the noise εi,MWj To eliminate this, the Melbourne-Wübbena linear combination for each satellite is reduced in some embodiments by the satellite MW bias. bMWj for this satellite and smoothed over time (for example, averaged). Individual differentiation between satellites eliminates the receiver MW bias b. i,MW, whereby only one differentiated Widelane ambiguity expression remains per satellite-receiver-satellite link. The differentiated Widelane ambiguities are calculated using the (constant) Widelane wavelength λ. WL Since only single-differentiated Widelane ambiguities are used in the phase-time signal processor, this method is used in some embodiments as a substitute for using the Widelane ambiguities 340 from the MW bias processor 325 and / or for a quality check of the Widelane ambiguities 340 received by the MW bias processor 325. Part 9.2 Individually differentiated phase-adapted time signals

[0226] The location of each reference station is precisely known. The precise location of each satellite at each epoch is given by the current orbital position / velocity data 350 from orbital processor 330. The geometric distance ρij The tropospheric delay between a satellite j and a reference station i at each epoch is calculated from their known positions. Tij The code-time signal processor 320 receives data for each reference station.

[0227] The ionosphere-free linear combination Φi,IFj=ρij+cΔtΦ,i−cΔtΦj+Tij+λIFNi,IFj+εΦ,i,IFj is reduced by the (known) geometric distance ρij and the (known) tropospheric delay Tij, where the only unknown is the ionosphere-free ambiguity expression λIFNi,IFj=λNLNi,1j+12(λWL−λNL)Ni,WLj the satellite phase-time signal error expression cΔtΦj:=cΔt+bΦ,IFj and the receiver phase-time signal error expression cΔt Φ,i := cΔt+b Φ,i,IF left over.

[0228] Differentiating the observations of the two satellites individually on the same receiver eliminates the receiver time signal error.

[0229] Reducing this single difference in the corresponding single-difference wide-lane ambiguity leads to the single-difference phase-time signal together with a single-difference N ambiguity (often also called narrow-lane ambiguity in this context, since the corresponding wavelength here is λ). NL is). Φi,IFj1j2−ρij1,j2−Tij1j2−12(λWL−λNL)Ni,WLj1j2=−cΔtΦj1j2+λNLNi1j1j2+εΦ,i,IFj1j2

[0230] This is calculated for each station observing the same pair of satellites. At this point, it is impossible to distinguish between the single-difference satellite time signal error and the single-difference narrow-lane ambiguity expression, which is an integer multiple of the narrow-lane wavelength.

[0231] If the single-difference ambiguity is set to zero, a single-difference phase-time signal is generated. cΔt˜Φ,ij1j2:=−(Φi,IFj1j2−ρij1,j2−Tij1j2−12(λWL−λNL)Ni,WLj1j2) The phase time signal is shifted by an integer number of narrowlane cycles. This phase time signal does not have a fixed narrowlane state. The difference between two of these single-difference time signals cΔt˜Φ,i1j1j2 and cΔt˜Φ,i2j1j2 is an integer number of narrowlane cycles. Part 9.3 Smoothed Individually Differentiated Phase-Time Signals

[0232] For each pair of satellites, the individually differentiated phase-time signal errors observed by different stations are shifted to a common level and averaged using fixed narrowlane ambiguities to obtain a more precise time signal error: cΔt¯Φj1j2=1n∑i=1n(cΔt˜Φ,ij1j2+λNLNj1j1j2) Part 9.4 Phase-Adapted Time Signal Estimation Part 9.4.1 Spanning Tree-Based Phase Time Signals

[0233] Some embodiments use a spanning tree approach to estimate phase-matched time signals. To calculate a single-difference time signal error between any pair of satellites, a set of single-difference phase-time signals is needed for the satellite-to-satellite links, such that there is a unique path to reach each satellite via satellite-to-satellite links, starting from a dedicated or specified reference satellite. If all satellites are nodes of a graph, such a path is called a spanning tree. If each edge of the graph is provided with a weight, a minimum spanning tree is a spanning tree with a minimum sum of edge weights. Some embodiments use a weighting scheme-based discrete category for an edge between two satellites and assign the following phase-time signal values ​​to the edge: i. Several edges connecting satellites j1 and j2 have a fixed N1 ambiguity: (weighted) averaged single-differentiated time signal cΔ¯tΦj1j2, ii. only a single edge connecting satellites j1 and j2 has a fixed N1 ambiguity: cΔt˜Φ,ij1j2+λNLNi1j1j2, iii. no edge connecting satellites j1 and j2 has a fixed N1 ambiguity: cΔt˜Φ,ij1j2 For receiver i with min(Elevation (j1,j2) is maximum, iv. No WL ambiguity available for an edge connecting satellites j1 and j2: Do not use such an edge; therefore, no phase-time signal needs to be defined.

[0234] Each edge of category (i) has a lower weight than an edge of category (ii), each edge of category (ii) has a lower weight than an edge of category (iii), etc. Edges within each category have a continuous weighting, which in some embodiments is derived in category (i) from the variance of the average or mean, and in categories (ii) and (iii) from the elevations or heights at which the satellites are seen in the single difference at the corresponding station.

[0235] If the minimum spanning tree uses an edge without a fixed narrowlane state, the narrowlane ambiguity will occur. Ni1j1j2 Set to zero, a fixed state is reached. Selecting a reference satellite with phase-time signal error. cΔtΦjref, Set to zero, individually differentiated phase-time signal errors are determined. cΔtΦjref1i2 The phase-time signal error of the satellite is calculated for all other satellites, solving a linear system. cΔtΦj2:=cΔtΦjrefj2 Part 9.4.2 Filter estimation of phase-matched time signals

[0236] Some implementations use a filter approach to estimate phase-matched time signals or clocks. A filter is set up with all satellite phase-time signal errors as states. The state of the reference satellite's time signal error cΔtΦjref is set to zero. In some embodiments, all connections from the spanning tree and additionally all connections with fixed narrowlanes are added to the filter for more precise estimation of individual differentiated phase-time signal errors. Part 9.5 Narrowlane filter bank

[0237] Once a set of individual differentiated phase-time signal errors has been estimated, for example in Part 9.4, all observations for shifted individual differentiated phase-time signals in Part 9.2 are reduced by the time signal errors estimated in Part 9.4: Ni1j1j2≈1λNL(cΔt¯Φj1j2−cΔt˜Φ,ij1j2)

[0238] What remains is an observable for the integer narrowlane offset. For each station, a narrowlane filter with one narrowlane ambiguity state per satellite is updated with these observations. Part 9.6 Individually differentiated code-adapted time signals at high rate

[0239] The Phase Time Signal Processor 335 also calculates code-adapted time signals or clocks at a high rate. Part 9.6.1 Intermediate storage of low-rate code-adapted time signals

[0240] GNSS observations (reference station data 305 or 315) for a given time (for example, an epoch t1) are buffered for use when low-rate information arrives at the same time (t1) from the code-matched time signal processor 63. This information includes two of the following: time signal errors 365, tropospheric delays 370, and ionospheric-free glide ambiguities 374. Therefore, GNSS observations that coincide in time with the low-rate time signal processor information will always be available from an observation buffer. When a set of code-matched low-rate time signals 365 arrives, they are combined with the GNSS observations, the tropospheric delays 370, and time-matched satellite position / velocity information 350 to calculate the carrier ambiguities.

[0241] Fig.Figure 27 shows the data flow at 2700 for the case where the satellite time signals (time signal errors) 365 and the tropospheric delays 370 are used. GNSS observations arrive over time at the phase-time signal processor 335 with (for example, epoch) time tags t0, t1, t2, t3, etc. Code-adapted time signals at a low rate. cΔtPi Tropospheric delays T arrive asynchronously over time with time tags t0, t1, t2, t3, etc. An observation buffer 2705 holds the observations. These are combined for each time t. i in a 2710 combiner with satellite position / speed data, with code-adapted low-rate time signals cΔtPj and with the tropospheric delays T to generate the ionospheric-free glide ambiguity expressions λNij:=λIFNi,IFj+bΦ,i,IF−bP,i,IF−(bΦ,IFj−bP,IFj). A process 2715 calculates the individually differentiated code-adapted time signals 2720.

[0242] If the processor sends the floating ambiguities at a low rate, buffering or temporary storage of the observations is not required. The floating ambiguities can be used directly for future observations. Part 9.6.2 Calculating iono-free floating ambiguities in the phase-time signal processor

[0243] The data combiner (2710) forms an ionosphere-free linear combination of the carrier phase observation and reduces it by the geometric distance. ρij (calculated using the receiver and satellite position), the tropospheric delay Tij and the code-adapted satellite time signal error at a low rate cΔtPj (for example, 365-degree time signals). After this reduction, the receiver time signal error and a floating ambiguity remain. If this is done using inter-satellite single-differentiated observations, the receiver time signal is eliminated, and therefore only the single-differentiated, ionospheric-free ambiguity expression remains: Φi,IFj1j2(t1)−ρij1j2(t1)−Tij1j2(t1)+cΔtPj1j2(t1)=λNij1j2(t1)+εΦ,i,IFj1j2(t1) with cΔtPj1j2:=cΔtj1j2+bP,IFj1j2 and gliding ambiguity λNij1j2:=λIFNi,IFj1j2−(bΦ,IFj1j2−bP,IFj1j2), which is a constant value and is retained for use until the next update of the time signals at a low rate.

[0244] As an alternative to calculating the ionospheric-free ambiguities in the phase-time signal processor 335, the ionospheric-free floating ambiguities are obtained from a previous processor, such as ionospheric-free floating ambiguities 374 from the low-rate code-time signal processor 320. Part 9.6.3 Using iono-free floating ambiguities to compute high-rate code-time signals

[0245] Once iono-free ambiguities are known for time t1, single-differentiated ionosphere-free linear combinations can be used at any time in the future (for example, (t2) for each pair of satellites together with an ionospheric delay and present geometric distance; Φi,IFj1j2(t2)−ρij1j2(t2)−Tij1j2(t1)−λNij1j2(t1)=cΔtPj1j2(t2)+εΦ,i,IFj1j2(t2)+εΦ,i,IFj1j2(t1)

[0246] The result of this calculation is the individually differentiated, code-adapted satellite time signal error. cΔtPj1j2. This makes it possible to detect individually differentiated, code-adapted time signal errors at a high rate. cΔtPjrefj2 to estimate between a given satellite j2 and a selected reference satellite j ref If only the inter-satellite differentiated time signal errors are of interest, the reference satellite time signal error will be calculated. cΔtPjrej Set to zero, the reference time signal can be directed to the corresponding transmission or a low-rate time signal to shift the time signals to an absolute level. For simple representation, included Fig. 28A to 28C do not have this shift.

[0247] Fig.Figure 28A shows a first embodiment 2800 for preparing the code-matched high-rate satellite time signals 375. Precise satellite orbit information, for example satellite position / velocity data 350 or 360, and GNSS observations, for example reference station data 305 or 315, are supplied as inputs to a first process 2805 and as inputs to a second processor 2810. The first process 2805 estimates a set of ionospheric-free sliding ambiguities 2815. These are supplied to the second process 2810, which estimates the code-matched high-rate (time signal error) satellite time signals 375.

[0248] Fig.Figure 28B shows a second embodiment 2820 for preparing the code-matched high-rate satellite time signals 375. Precise satellite orbit information, for example, satellite position / velocity data 350 or 360, and GNSS observations, for example, reference station data 305 or 315, are supplied as inputs to a code-matched low-rate time signal processor 2825, for example, code-time signal processor 320, and as inputs to a code-matched high-rate time signal processor 2830. The code-matched low-rate time signal processor 2825 prepares ionospheric-free glide ambiguities 2835, for example, ionospheric-free glide ambiguities 374, and tropospheric delays 2840, for example, tropospheric delays 370.The ionospheric-free glide ambiguities 2835 and the tropospheric delays 2840 are fed to the code-matched high-rate time signal processor 2830, for example, forming part of the phase-time signal processor 335. The code-matched high-rate time signal processor 2830 uses the inputs at 2845 to compute individually differentiated code-matched high-rate satellite time signals 2850 per station. A filter 2855 uses these to qualitatively estimate improved code-matched satellite time signals (time signal errors) 2860, for example, code-matched high-rate satellite signals 375.

[0249] Fig.Reference 28C shows a second embodiment 2865 for preparing the code-matched high-rate satellite time signals 375. Precise satellite orbit information, for example satellite position / velocity data 350 or 360, and GNSS observations, for example reference station data 305 or 315, are supplied as inputs to a code-matched low-rate time signal processor 2870, for example code-time signal processor 320, and as inputs to a code-matched high-rate time signal processor 2875. The code-matched low-rate time signal processor 2870 prepares code-matched low-rate time signals 2880, for example code-matched low-rate satellite time signals 365, and tropospheric delays 2882, for example tropospheric delays 370.The low-rate code-matched satellite time signals 2880 and the tropospheric delays 2882 are fed to the high-rate code-matched time signal processor 2875, for example, forming part of the phase-time signal processor 235. The high-rate code-matched time signal processor 2884 uses the inputs at 2884 to compute ionospheric-free floating ambiguities 2886, which are used at 2888 to compute single-differentiated high-rate code-matched satellite time signals 2890. A filter 2892 uses these to qualitatively estimate improved high-rate code-matched satellite time signals (time signal errors) 2894, for example, high-rate code-matched satellite time signals 375. Part 9.6.4 Time signal shifter or clock shifter

[0250] Individually differentiated phase-adapted time signal errors cΔtΦjrefj2 and individually differentiated code-adapted time signal errors cΔtPjrefj2 The time signals are estimated as in the course of estimating phase-matched satellite time signals and single-differentiated code-matched satellite time signals at a high rate. The single-differentiated code-matched satellite time signals at a high rate exhibit the same accuracy as single differences of the code-matched satellite time signals at a low rate. The phase-matched satellite time signals are constructed to maintain the integer nature of the narrow-lane ambiguity when used for single-differentiated ionospheric-free carrier phase observations, along with the precise satellite orbits and the wide-lane ambiguities derived from the MW biases used in the time signal estimation. Therefore, the quality of a single-differentiated phase-matched time signal error is not altered if this time signal error is shifted by an integer of narrow-lane cycles.Since the phase-matched satellite time signal errors are always used in single-differentiation operation, such a shift applied to all satellite time signal errors is canceled out in single-differentiation operation. According to some embodiments, the following shift is applied to the phase-matched satellite time signal errors to maintain their values ​​close to the code-matched low-rate satellite time signal errors and to reduce their noise.

[0251] Therefore, in some embodiments, a whole number of narrowlane cycles are used. sNLjrefj2=Round(cΔtPjreƒj2−cΔtΦjrefj2) added to each of the phase-matched satellite signal errors to minimize the distance to the code-matched high-rate satellite time signals.

[0252] In some embodiments, the code-adapted low-rate time signal errors are approximated by a continuously controlled time signal. cΔtP,steeredj. The value of this controlled time signal from the reference satellite sref=cΔtP,neeredjref This is then added to all high-rate time signals. This makes all time signals close to the code-matched low-rate time signals, but the reference time signal is smooth.

[0253] In some embodiments, the mean of the difference between the phase-matched, high-rate satellite time signals and their corresponding controlled, low-rate time signal is calculated and added to all shifted, phase-matched satellite time signal errors. The same is done for code-matched, high-rate satellite time signals. This procedure introduces noise into the reference time signal or reference clock. If the number of time signals used to calculate this mean changes, the calculated mean may be discontinuous. To handle these jumps, one can simply send a jump message or smooth the mean, for example, by controlling it to the current value or using a moving average, etc. sP,noise=1n∑j=1n(cΔtPjrefj2−cΔtP,steeredjrefj2)sΦ,noise=1n∑j=1n(cΔtΦjrefj2−cΔtP,steeredjrefj2+sNLjrefj2)

[0254] The shifted time signal errors are cΔtPj2=cΔtPjrefj2+sP,noise+srefcΔtΦj2=cΔtΦjrefj2+sΦ,noise+sref+sNLjrefj2

[0255] The shift is primarily performed to keep the phase-matched satellite time signal errors close to GPS time, and therefore certain transmission methods are applicable. Additionally, the time signal bias can be adjusted. bΔij=cΔtPj−cΔtΦj which is the difference between the phase-matched satellite time signals and the code-matched satellite time signals, kept within a certain range. Part 9.6.5 Jump News

[0256] If the time signal bias deviates from its range, or if the satellite-to-satellite links of the phase-matched time signals were estimated without using fixed narrow-lane ambiguities, the offset will change and a jump message will be sent. This also means that the phase-matched satellite time signal error will change its level. Part 9.7 Phase-Time Signal Processor Implementation Forms

[0257] Fig.Figure 29 shows an architecture 2900 of a phase-time signal processor 335 according to some embodiments of the invention. The inputs to the phase-time signal processor are: MW biases and / or fixed WL ambiguities 2095 (for example, MW biases 345 and / or fixed WL ambiguities 340) or code-matched low-rate (LR) time signals 2910 (one per satellite, for example, code-matched low-rate satellite time signals 365), satellite orbit information (for example, precise satellite orbit positions / velocities 350 and / or IGU orbit data 360), tropospheric delays 2920 (one per reference station, for example, tropospheric delays 370), and GNSS observations 2925 (from multiple satellites at multiple reference stations, for example, reference station data 305). Choosing between MW bias or fixed WL ambiguities gives two options.One option is to use the low-rate MW bias, along with the low-rate orbit information 2915 and the GNSS observations 2925, in an optional fix bank 2930 to determine the WL ambiguities. These are then provided at a higher rate (HR) as differentiated fixed WL ambiguities 2935 to a computation process 2940. A second option is to use the low-rate fixed WL ambiguities directly to provide differentiated (SD) fixed WL ambiguities to process 2940. High rate means that the differentiated fixed WL ambiguities used in the high-rate process 2940 remain the same from epoch to epoch in process 2940 between low-rate updates.

[0258] The low-rate code-matched time signals 2910 are used in process 2945 to calculate single-differentiated, low-rate ionospheric-free glide ambiguities 2950. These are used with the low-rate orbit data 2915, the low-rate tropospheric delays 2920, and the high-rate GNSS observations 2925 in a process 2955 to calculate single-differentiated, high-rate code-matched time signals 2960 (one per satellite, such as high-rate code-matched satellite time signals 375).

[0259] The single-differentiated fixed WL ambiguities at high rate 2935 are used with the low-rate orbital data 2915, the low-rate tropospheric delays 2920, and the GNSS observations 2925 in a process 2940 that computes phase-matched time signals at high rate 2945 (one per satellite, for example, phase-matched satellite time signals at high rate 375). The phase-matched time signals at high rate 2945 are used together with the orbital data 2915, the tropospheric delays 2920, and the GNSS observations 2925 in an ambiguity determination bank 2975 that attempts to determine single-differentiated ambiguities, for example, L1 ambiguities. The individually differentiated fixed ambiguities 2980 are pushed into the process 2965 to help calculate phase-matched time signals at a high rate 2970.

[0260] The MW bias and / or fixed WL ambiguities 2905 and the code-matched low-rate time signals 2910 and the individually differentiated code-matched high-rate time signals 2960 and phase-matched high-rate time signals 2970 are fed into a process 2985, which shifts and combines them to provide a high-rate data stream 2990 containing at least phase-matched high-rate satellite time signals, code-matched high-rate time signals, and MW bias at a high rate. Data 2990 is fed to the scheduler 355, as described in Fig. 3 described.

[0261] Fig.Figure 20A shows an embodiment 3000 of a process for estimating phase-matched satellite time signals at a high rate. Precise satellite orbit information, for example, satellite position / velocity data 350 or 360, and GNSS observations, for example, reference station data 305 or 315, are fed at a low rate as inputs to a first process 3015 and at a high rate as inputs to a second process 3020. The first process 3015 estimates ambiguities 3025, one set of ambiguities per receiver. Each ambiguity corresponds to one from a receiver-satellite link and one from a satellite-receiver-satellite link. These ambiguities are fed at a low rate to the second process 3020, which estimates the phase-matched time signals at a high rate 3030.

[0262] In general, a linear combination exhibits receiver-dependent parameters p prior to carrier-phase observations. ron, such as receiver positions, receiver time signal errors or receiver bias, satellite-dependent parameters p s , such as the satellite position, as well as receiver-satellite connection-dependent parameters prs, such as tropospheric or ionospheric retardation, Ωrks,k∈L represent linear combinations of code and carrier phase observations with wavelengths λ k and ambiguities Nrks. A code and carrier phase combination as assumed here can be described as Ω^rs:=∑k∈iakΩrks=ƒ(pr,ps,prs)+cΔts+∑k∈L^ak(bks+λkNrks)+εrs with L̂⊆L, L̂ non-empty and real factors a k ∈ ℜ.

[0263] It should be noted that f: ℜ np→ℜ is linear in most of the parameters. If a mapping exists to convert between a receiver-satellite link-dependent parameter and a receiver-dependent parameter, it can be used to reduce the number of unknowns. For example, tropospheric delay can be represented by a model as a zenith delay mapped onto a line of sight. Therefore, instead of having one parameter per receiver-satellite link, only one parameter per receiver is needed.

[0264] Using special linear combinations, such as an ionospheric-free combination, parameters can cancel each other out. Any additional input of at least one of the parameters contained in p r , p s and prs It simplifies and accelerates the estimation process. In satellite-receiver-satellite individual differences, the p values ​​cancel each other out. rThey cancel each other out. Everything that follows can be done with single or zero differences, but this is not explicitly noted in every step.

[0265] If all parameters p r , p s , p s , and cΔt s If the known parameters and the noise are neglected, the remaining part is... ∑k∈L^ak(bks+λkNrks) Not unambiguous without additional information. Setting the ambiguities for a specific satellite-receiver combination to zero will shift the bias accordingly. If the ambiguities are known, the bias level is defined; if the bias is known, the level of ambiguities is defined.

[0266] In some linear combinations Ωrks The parameters of the code and carrier phase observations stand out. pr,ps,prs and cΔt s on, as is also the case with receiver biases b rkThis allows for an estimation of the biases. bks, brk and the ambiguities Nrks, separate from the other parameter.

[0267] Not all satellite biases can be separated from the satellite time signal error cΔt. s The error can be estimated. In this case, the sum of the true satellite time signals and the bias are estimated together, and the result is simply called the satellite time signal error. In this case, shifting a bias is equivalent to shifting a time signal.

[0268] A distinction is made below between ambiguities related to biases and ambiguities related to time signal errors. The time signal ensemble, and also the bias ensemble, estimated from GNSS observations, are underdetermined. Therefore, one of the satellite or receiver time signal errors / biases, or a combination thereof, can be set to any desired value or function to make the system solvable. In some embodiments, one of the time signals / biases is set to zero, or additional measurements or conditions are added to one of the time signal errors / biases or a linear combination of time signal errors / biases. The examples given here always include a reference time signal / bias, which is also set to zero, for illustrative purposes, but this is not a requirement, and another approach can be used.

[0269] Process I (3015): In a first step, all other input parameters are used to reduce the linear combination of the carrier-phase observations. Depending on the linear combination used, there are small differences in how phase-matched satellite time-signal errors are estimated. One option is to use a single large filter: in this case, all remaining unknown parameters of p are removed. r , p s and prs, the satellite time signal errors, including biases, and all ambiguities of Ω^rs This is represented by a model as states in a filter, for example, a Kalman filter. For example, the filter used for orbit determination in Part 8 (Orbit Processor), the one used for code-matched time signals in Part 6 (Code-Matched Time Signal Processor) (except for the integer nature of the ambiguities), or the Melbourne-Wübbena bias processor in Part 7 (MW Bias Processor) can be used. Another option is to perform hierarchical estimation.

[0270] In some embodiments, linear combinations can be used. Ωrks from code and carrier phase observations, in which the parameters pr,ps,prs and cΔt s cancel out the biases and ambiguities of Ωrks, estimated in an auxiliary filter, used to simplify the main filter.

[0271] Another option is to use a bank of filters instead of a single filter, or a main filter with auxiliary filters. If all parameters are known, along with any ambiguities, then at least one of the code and carrier-phase combinations can be used. Ω^rs or Ωrks,k∈L^ or the ambiguities can be estimated in a filter bank with one filter for each ambiguity or one filter for each receiver. This is done for single-differentiated observations in the phase-time signal processor to estimate the Widelane ambiguities using the Melbourne-Wübbena linear combination and Melbourne-Wübbena bias as input.

[0272] Another option is to use a combination of a main filter with a bank of filters. In some embodiments, the results from the filter banks are used to reduce the carrier-phase combinations. Φ^rs in the main filter, which is used to estimate the remaining unknowns.

[0273] At least one set of fixed ambiguities is sent from Process I (3015) to Process II (3020), but all of the estimated parameters could also be sent.

[0274] Process II (3025): In Process II, the linear combination Ω^rs The error is reduced by all input parameters of Process I or additional sources. If no fixed ambiguities are available for the Process I ambiguities that contribute to the time signal error, a subset of such ambiguities, defined, for example, by a spanning tree, is set to an arbitrary integer in some embodiments and used like fixed ambiguities, as discussed above. If this subset changes or is replaced by fixed Process I ambiguities, the resulting satellite phase time signal errors may change their level. All remaining unknowns are represented by a model as states in a filter, for example, a Kalman filter.

[0275] As in Fig.In some embodiments, 30A estimates the ambiguities at a first rate and estimates a phase-matched time signal per satellite at a second rate higher than the first rate. The ambiguities estimated in Process I (3005) are constant as long as the receiver does not exhibit cycle slippage. Therefore, an estimated ambiguity can be used for a long time. Also, some of the other conditions, such as the troposphere estimated in Process I (3005), vary slowly and can be assumed to be constant for a time. If the observations of Process II (3020) are reduced by such constant and slowly varying parameters, the filter used to estimate time signal errors has few unknowns remaining and is therefore quite fast. In general, Process II can operate at a higher update rate than Process I.

[0276] Fig.Figure 31 shows an embodiment 3100 in which estimating phase-matched time signals per satellite comprises: using at least the satellite orbit information, the ambiguities, and the GNSS signal data to estimate a set of phase-matched time signals per receiver, each phase-matched time signal corresponding to one from a receiver-satellite link and a satellite-receiver-satellite link; and using a plurality of the phase-matched time signals to estimate one phase-matched time signal per satellite. The satellite orbit information 3105 (for example, 350 or 360) and the GNSS observations 3110 (for example, 305 or 315) are used in a first process 3115 to determine the ambiguities 3120.The ambiguities 3120 are used with the satellite orbit information 3105 and the GNSS observations 3110 in a second process 3125 to estimate the set of phase-matched time signals 3130, where each phase-matched time signal corresponds to one from a receiver-satellite link and one from a satellite-receiver-satellite link. These are used in a third process 3135 to estimate satellite time signals 3140, one per satellite.

[0277] This means that instead of having a large filter in the second process, the problem can be decoupled by using a small filter for each satellite-to-satellite link to estimate time signal errors per receiver-to-satellite link, or in a single-difference case, per satellite-to-receiver-to-satellite link. These time signal errors per link can then be combined, either using only links defined by a spanning tree (for example, as in 3058) or by using a filter with one time signal error state per satellite.

[0278] In some embodiments, the ambiguities are estimated using at least one previously estimated phase-matched time signal per satellite. As noted above, the known fixed ambiguities define the time signal error level, and (conversely) a known time signal error leads to ambiguities that fit this time signal error. Therefore, the feedback of time signal error estimates back to Process I allows ambiguities to be estimated without having a specific time signal error state. Since Process II can generate time signal error estimates before all ambiguities are fixed, this feedback is advantageous in such a scenario for fixing ambiguities that correspond to time signal errors. Fig.Figure 72 shows such an embodiment 3200. The satellite orbit information 3205 (for example, 350 or 360) and the GNSS observations 3210 (for example, 305 or 315) are used in a first process 3215 to determine the ambiguities 3220. The ambiguities 3220 are used with the satellite orbit information 3205 and the GNSS observations 3210 in a second process 3225 to estimate the set of phase-matched time signals 3230. These are fed back to the first process 3215, where they are used in estimating the ambiguities 3220.

[0279] Two more detailed implementations of this process are presented in Fig. 30B and Fig. 30°C.

[0280] Fig. Figure 30B shows a simplified schematic diagram of an embodiment of an alternative phase-time signal processor 3035 with WL ambiguities input as in Option II of Fig.29 (compare with process 2965). Satellite orbit data 3005 (e.g., 350 or 360), GNSS observations 3070 (e.g., 305 or 315), tropospheric delays 3040 (e.g., 370), and fixed WL ambiguities 3045 (e.g., 340) are fed to a filter bank 3050 of process I (3015). The filter bank 3050 estimates single-differentiated free narrowlane ambiguities 3055. The SD-free NL ambiguities 3055 and the fixed WL ambiguities 3045 are combined in a process 3060 with the satellite orbit data 3005, the GNSS observations, and the tropospheric delays 3040 to calculate single-differentiated satellite time signals. 3062. These are applied to a narrow-lane filter bank 3064 to estimate individually differentiated satellite time signals 3066. A spanning-tree process 3068 (for example, MST) is applied to these to generate a set of individually differentiated satellite time signals 3070.These are fed back to filter bank 3050 from process I (3015) to improve the estimation of single-differentiated free narrowlane ambiguities 3055.

[0281] Fig. Figure 30C shows a simplified schematic diagram of an embodiment of an alternative phase-time signal processor 3075 with MW biases input as in Option I of Fig. 29 (compare with process 2965). Satellite orbit data 3005 (e.g., 350 or 360), GNSS observations 3010 (e.g., 205 or 315), tropospheric delays 3040 (e.g., 370), and fixed WL ambiguities 3045 (e.g., 340) are fed to a filter bank 3078 from process I (3015). Filter bank 3078 estimates single-differentiated free narrowlane ambiguities 3088. A filter bank 3082 uses the MW satellite biases 3045 to estimate WL ambiguities 3084. As in Fig.In process 30B, the SD-free NL ambiguities 3055 and the fixed WL ambiguities 3084 are combined in a process 3060 with the satellite orbit data 3005, the GNSS observations 3010, and the tropospheric delays 3040 to calculate single-differentiated satellite time signals 3062. These are applied to a narrow-lane filter bank 3064 to estimate single-differentiated satellite time signals 3066. A spanning-tree process 3068 (for example, MST) is applied to these to generate a set of single-differentiated satellite time signals 3070. These are fed back to the filter bank 3078 of process I (3015) to improve the estimation of the single-differentiated free narrow-lane ambiguities 2080.

[0282] Using time signal error estimates from a second phase time signal processor, as described in Process I, allows estimating ambiguities that fit these time signals and ultimately enabling a fit, and ultimately estimating satellite time signal errors at the same level as the other time signal errors of the processor. For this embodiment, it is advantageous to have a secondary time signal processor as a replacement or backup, available to immediately provide time signal error estimates without a level change in the event of a failure of the primary time signal processor.

[0283] In some embodiments, at least one additional phase-matched time signal per estimated satellite is obtained from an external source and used to estimate the ambiguities. Fig.Figure 33 shows such an embodiment 3300. In part 3335, the satellite orbit information 3305 (for example, 350 or 360) and the GNSS observations 3010 (for example, 305 or 315) are used in a first process 3315 to determine the ambiguities 3320. The ambiguities 3320 are used with the satellite orbit information 3305 and the GNSS observations 3310 in a second process 3325 to estimate the set of phase-matched time signals 3330. In part 3385, the satellite orbit information 3355 (for example, 350 or 360) and the GNSS observations 3060 (for example, 305 or 315) are used with one or more of the satellite time signals 3330 in a first process 3315. Process 3365 for estimating the ambiguities 3370. In this embodiment, part 3335 is an external source of the satellite time signals 3330 with reference to part 3385.The ambiguities 3370 are used with the satellite orbit information 3355 and the GNSS observations 3360 in a second process 3375 to estimate the set of phase-matched time signals 3380.

[0284] In some embodiments, at least one set of ambiguities per receiver is determined for additional receivers, where each ambiguity corresponds to one of a receiver-satellite link and one of a satellite-receiver-satellite link. After determining the ambiguities for the additional receivers, at least the precise orbit information, the ambiguities for the additional receivers, and the GNSS signal data are used to estimate the at least one additional phase-matched time signal per satellite.

[0285] Fig.Figure 34 shows such an embodiment 3400. In part 3455, the satellite orbit information 3405 (for example, 350 or 360) and the GNSS observations 3310 (for example, 305 or 315) are used in a first process 3415 to determine the ambiguities 3420. The ambiguities 3420 are used with the satellite orbit information 3405 and the GNSS observations 3410 in a second process 3425 to estimate the set of phase-matched time signals 3430. In part 3485, the satellite orbit information 3455 (for example, 350 or 360, but optionally from a different orbit processor, associated with a different network of reference stations) and the GNSS observations 3410 (for example, 305 or 315, but optionally from a different network of reference stations) used with one or more of the satellite time signals 3430 in a first process 3465 to estimate the ambiguities 3470.The ambiguities 3470 are used with the satellite orbit information 3455 and the GNSS observations 3460 in a second process 3475 to estimate the set of phase-matched time signals 3480.

[0286] The primary and secondary timing signal processors can also be of different types. This option can be used to estimate ambiguities close to the level of timing signal errors based on a different linear combination (for example, code timing signal errors or different phase combinations). Using such ambiguities in Process II results in timing signal errors close to those input in Process I. Part 10: Scheduler & Message Encoder

[0287] The Scheduler 355 and Message Encoder 385 can be implemented as desired. Methods and devices for encoding and transmitting satellite information are described, for example, in patent applications US2009 / 0179792 A1, US2009 / 0179793 A1, US2010 / 0085248 A1, US2010 / 0085249 A1 and / or US2010 / 0214166 A1. Part 11: Rover Processing with Synthesized Reference Station Data Part 11.1 Introduction

[0288] Existing RTK rover positioning engines are typically designed to process differentiated data; the rover data is differentiated with base station data, and filters operate on the differentiated data. GNSS observations are contaminated with a number of errors, such as satellite time signal errors, receiver time signal errors, tropospheric delay errors, and ionospheric delay errors. Satellite-related errors (e.g., satellite time signal errors) can be eliminated if the difference between the observations of two receivers observing the same satellite is used. If such receivers are positioned close enough to each other (e.g., a few kilometers under normal conditions), then atmospheric-related errors can also be eliminated.In the case of VRS (Virtual Reference Station), the difference is not taken between two stations, but between the rover station and a virtual station whose data is generated using observations from a network of receivers. This network makes it possible to generate knowledge about how faults behave across the network's region, thus enabling differential positioning over longer distances.

[0289] Previous approaches to Precise Point Positioning (PPP) and PPP with Ambiguity Resolution (PPP / RTK) remove model errors by applying them as corrections (subtracting the errors) to the rover data. While this works, using a rover receiver configured to process differentiated data requires a change in data processing, replacing single-differentiation with rover data-only error correction before the data can be processed.

[0290] This implies two different operating modes within the rover's positioning engine. In practice, this results in separate processors for PPP and RTK. This consumes significant software development resources and occupies more of the rover's CPU memory for additional modules and data. Part 11.2 Global Virtual Reference Station Positioning

[0291] Some embodiments of the invention are based on a substantially different approach in which a synthesized base station (SBS) data stream is generated for any position on or near the Earth's surface using precise satellite information (for example, precise orbits and time signals). This data stream is equivalent to having a real reference station near the rover.

[0292] For processing in the rover or mobile unit, a real-time kinematic (RTK) positioning engine is used, which is designed to handle the different error characteristics of PPP compared to traditional virtual reference station (VRS) processing. Unlike traditional VRS processing, some embodiments of the invention use a processing approach that is not based on small ionospheric residues. And unlike traditional VRS processing, some embodiments of the invention optionally process different pseudo-distance observables.

[0293] PPP and PPP / RTK functionality is retained with comparatively little software development and few changes to the rover positioning engine, while maintaining the advantage of the well-proven RTK engines (virtual computer machines) that have been developed and perfected over many man-years of development time. Examples of such functionality include processing data collected in the sky and addressing delays in the reference data / corrections (low-latency positioning).

[0294] PPP-RTK studies employing the SBS technique have demonstrated the high performance of such a system. A horizontal positioning accuracy of 10 cm (95%) was achieved after approximately 600 seconds (average) when processing a test dataset. Convergence to the typical long baseline and VRS survey accuracy of 2.54 cm horizontal (95%) were achieved after 900 seconds (average). This suggests that SBS techniques described herein can provide sub-inch horizontal positioning performance. Part 11.3 Generating SBS Data

[0295] SBS technology enables the generation of virtual GNSS data (data from a virtual GNSS reference / base station) for any position on or near the Earth's surface using precise satellite information (e.g., orbits, time signals). The core processing takes place within a module responsible for generating virtual data. The goal of such data generation is to allow the use of precise GNSS satellite information in a GNSS receiver running a conventional real-time kinematic (RTK) process, as is typically used with reference receiver data from a physical base station or a virtual reference station. In SBS technology, the receiver's antenna position is calculated by the RTK engine using differential GNSS data processing (i.e.,The difference between observations from a reference receiver and a rover receiver using SBS data as provided by the (virtual) reference receiver, and the rover receiver data. This technique therefore allows a differential GNSS processor to be used to calculate positions anywhere without the explicit need for a nearby reference station.

[0296] The SBS module uses at least one of the following: Phase-matched time signals: These are the satellite time signal offsets, calculated as described in Part 9 (phase-matched time signal processor). Code-adapted time signals: These are the satellite time signal offsets, as described in Part 6 (code-adapted time signal processor). Melbourne-Wübbena biases: These are the satellite biases for the Melbourne-Wübbena phase and code combination, calculated as described in Part 7 (MW bias processor). Jump messages: These messages can indicate whether or not the satellite phase-time signals have undergone a level change in the recent past (e.g., 10 minutes). The reasons for a level change are shown in Part 9 (phase-matched time signal processor). Whenever a jump (level change) occurs in the satellite phase-time signal offset, action must be taken on the rover. Such an action might be resetting the satellite's ambiguity(ies) in the RTK engine. Approximate Rover Position: This is the position for which the virtual base data is generated. The approximate position of the rover can be used, for example, so that geometrically dependent components (such as satellite position) are the same as for the rover data. Time Tag: This is the time (epoch) for which the virtual data is generated. The virtual data is generated for certain points in time (epochs) that depend on the rover observation time tags, so that it can be used in differential data processing together with the rover data.

[0297] Given one or more of the elements from the list above as input, the SBS module generates a set of GNSS virtual observations as output. These observations can include, but are not limited to: L1 code, L2 code, L1 phase, L2 phase, L1 cycle slip information, L2 cycle slip information, and / or exact virtual base position. Once the virtual reference station dataset is available, it can be supplied to the RTK engine for differential processing with the rover's own GNSS observation data and optionally using the precise satellite data. The RTK engine can then apply conventional RTK processing to calculate the rover's coordinates. Part 11.4 Moving Base

[0298] The best corrections for kinematic rovers are obtained when the SBS position and the synthesized reference station data for the SBS position are frequently updated, for example, by generating a new set of SBS data for each epoch of rover observations. Some implementations use as the SBS position an initial estimate of the rover position, derived, for example, from a simple navigation solution using the rover observations.

[0299] In the case of state-of-the-art Virtual Reference Station (VRS) processing, a moving rover can cause a significant separation between the rover's position and the VRS location for which the VRS data is synthesized. Some implementations mitigate this by changing the VRS position when this distance exceeds a certain threshold. This can lead to restarts or resets of the RTK engine in many implementations.

[0300] For position determination (direction, wing control, etc.), typical RTK processing engines are usually able to process data from a moving base station; frequent updates of the SBS position and the synthesized reference station data for the SBS position do not require any modification of these rover processing engines. Part 11.5 SBS Implementation Forms

[0301] Fig. Figure 38 shows an embodiment 3800 of an SBS processor according to the invention. Rover receiver 3805 receives GNSS signals from several GNSS satellites, three of which are shown at 3810, 3815 and 3820. The receiver 3805 derives GNSS data 3825 from code observations and carrier-phase observations of the GNSS signal over several epochs.

[0302] Precise satellite data 3830 for the GNSS satellites are received, for example via a correction message 390 transmitted by a communications satellite 3835 or by other means, and decoded by a message decoder 3832. An SBS module 3835 receives the precise satellite data 3830 and also receives information which it can use as a virtual base location, such as an approximate rover position with time-tag 3842, generated by an optional navigation processor 3845. The approximate rover position is optionally obtained from other sources, as described below.

[0303] SBS module 3835 uses the precise satellite data 3838 and the approximate rover position with time tag 3842 to synthesize base station data 3850 for the virtual base location. The base station data 3850 includes, for example, synthesized observations of L1 code, L2 code, L1 carrier phase, and L2 carrier phase, and optionally includes information regarding L1 cycle shift, L2 cycle shift, and the virtual base location. The SBSD module 3835 is triggered by an event or the arrival of information indicating that a new epoch of synthesized base station data is to be generated. In some embodiments, the trigger is the availability of a rover observation epoch dataset. In other embodiments, the trigger is the current rover time tag. In some embodiments, an epoch of synthesized base station data 3850 is generated for each epoch of GNSS data observations 3825.In some embodiments, the trigger is the availability of an updated set of precise satellite data 3830.

[0304] In some embodiments, a differential processor 3855, such as a typical RTK positioning engine of an integrated GNSS receiver system 3700, receives the precise satellite data 3830, the synthesized base station data 3850, and the GNSS data 3825 from the rover receiver 3805 and uses them to determine a precise rover position 3860. Synthesized base station data 3850 are substituted for base station data in such processing.

[0305] Fig. Figure 39 shows a time signal prediction between an observation time Obs 0 and a subsequent observation time OBs 1.

[0306] Fig. Figure 40 shows a schematic block diagram of the SBS module 3835,

[0307] The SBS module 3835 uses at least one of the following: Phase-adjusted time signals 370: These are the satellite time signal offsets, calculated as described in Part 9 (Phase Time Signal Processor). Code-adapted time signals 365: These are the satellite time signal offsets, calculated as described in Part 6: Standard Time Signal Processor. Melbourne-Wübbena bias 345: These are the satellite biases for the Melbourne-Wübbena phase and code combination, calculated as described in Part 8: Widelane bias processor. Jump messages (for example, from Correction Message 390): Jump messages indicate when the satellite phase-time signal has undergone a level change in the recent past (for example, 10 minutes). The reasons for a level change are explained in Part 9: Phase-Time Signal Processor. When a jump (level change) is shown in the satellite phase-time signal offset, an action is taken on the rover, such as resetting the satellite's ambiguity(ies) in the RTK engine. Approximate rover position 3840: This is the position for which the virtual base data is generated. The approximate position of the rover can be used so that the geometrically dependent components (e.g., satellite position) are the same as for the rover data.

[0308] Time-Tag 3842: This is the time for which the virtual data is to be generated. The synthesized base station data 3850 are generated for certain time points that depend on the rover observation time-tags, so that they can be used in differential data processing together with the rover data.

[0309] When one or more of these elements are considered as input, the SBS module 3850 generates a set of GNSS virtual observations as output 3850. These observations include, but are not limited to: L1 code, L2 code, L1 phase, L2 phase, L1 cycle slip information, L2 cycle slip information, and exact virtual base position. The virtual base station is fed to the differential processor 3855, as are the GNSS data from the rover 3825 and, optionally, the precise satellites 3830. The differential processor 3855 calculates the coordinates of the precise rover position 3860, for example, using a conventional RTK receiver processing approach (see, for example, P. Misra et al., Global Positioning System: Signals, Measurements, and Performance, 2nd edition (2006), pages 239-241, and US Patent 5,610,614 granted on March 11, 1997, to Talbot et al.).

[0310] At any given time, the SBS module 3835 can receive one or more of the following: Approximate rover position 3840, precise satellite data 3830, and / or a time tag 3842. The approximate rover position 3840 is stored at 4005 as an updated current virtual base position. The precise satellite data 3840 is stored at 4010. The time tag or time label 3842 is stored at 4015. Any of these elements, or an optional external trigger 4020, can be used as a trigger event at a decision point 4025, initiating the generation of a new set of synthesized base station data 3850.

[0311] The satellite data for the current time-day are evaluated at 4030. This means that the stored satellite position and time errors are converted to the correct transmission time of the signal so that they can be used consistently with the rover observations. This is done because the stored satellite position and time signal errors do not necessarily match the time-day of each requested epoch of the SBS module. The precise satellite data set for the current time-day, 4035, is used at 4040 to synthesize a base station data set for the current time-day. At 4040, the satellite position and satellite velocity are calculated for the current time-day. The geometric distance between virtual base station i and satellite j is calculated, for example, as: ρij=((Xj−Xi)2+(Yj−Yi)2+(Zj−Zi)2) where X j ,Y j ,Zj the satellite position at the time of the current time-day is and X i ,Y i ,Z i The virtual base station location at the current time / day.

[0312] The neutral atmospheric (or troposphere) delay Tij is calculated, for example, using a predictive model. For examples of predictive models, see [Leandro 2009], [Leandro et al. 2006a], or [Leandro et al. 2006b].

[0313] The ionospheric delay Iij The Ll frequency is calculated, for example, using an ionospheric model. This can be a predictive model, such as the GPS transmission ionospheric model [ICD-GPS], or something more advanced. Optionally, the ionospheric delay can be set to zero.

[0314] The uncorrected synthesized base station data set for the time of the time day is then calculated, for example, as: Φi,1j'=ρij−cΔtΦj+Tij−Iij Φi,2j'=ρij−cΔtΦj+Tij−ƒ12ƒ22Iij Pi,1j'=ρij−cΔtPj+Tij+Iijƒ2ƒ1(λWL⋅bMWj−cΔtΦj+cΔtPj) Pi,2j'=ρij−cΔtPj+Tij+ƒ12ƒ22Iij+ƒ1ƒ2(λWL⋅bMWj−cΔtΦj+cΔtPj) where the synthesized L1 carrier observation for the virtual base site is, the synthesized L2 carrier observation for the virtual base site is, the synthesized L1 code observation for the virtual base location is, and The synthesized L2 code observation for the virtual base location is.

[0315] The uncorrected synthesized base station dataset 4045 is corrected at 4050 to produce a synthesized base station dataset 3850 for the current time day. The corrections include one or more of the effects described in Part 3: Observational Data Correctors, such as crustal tides, phase wind-up, and antenna phase center variation. The corrected synthesized base station dataset is: Φi,1j=Φi,1j'−Ci,1j Φi,2j=Φi,2j'−Ci,2j Pi,1j=Pi,1j'−Ci,1j Pi,2j=Pi,2j'−Ci,2j

[0316] The synthesized base station record generation is then completed at the current time day and delivered to the differential processor 3855.

[0317] In some embodiments, the 3855 differential processor uses the transmission ephemerides to determine the satellite position and the time signal error, since only approximate values ​​for the satellite are required in this positioning mode. This is also the case when the differential processor uses SBS data; however, in some embodiments, the processor optionally uses the available precise satellite data.

[0318] Fig. Figure 42 shows an alternative embodiment 4200, which is a variant of the processing 3800 of Fig. 38. In this embodiment, the precise satellite data 3830 and rover observation data 3825 are sent to a PPP (Precise Point Positioning) engine 4210 instead of or in addition to the differential processor 3855. The PPP engine 4210 provides the rover coordinates instead of or in addition to these from the differential processor 3855.

[0319] Fig.Figure 43 shows a simplified view of the embodiment of Fig. 38. The synthesized GNSS data 3850 are generated for a given location using precise satellite data 3830. The synthesized data 3850 are forwarded to the differential processor 3855, which, also using the rover GNSS data 3825, calculates the rover position 3860.

[0320] Fig. 44 shows a timing diagram of a low-latency version of the processing of Fig. 38, Fig. 42 or Fig.44. In this variant, the arrival of an epoch of rover observation data (for example, 3825) or an epoch time day (for example, 3842) is used as a trigger for the generation of synthesized base station data (for example, 3850). For example, a set of synthesized base station data (for example, 3850) is generated for each epoch of rover observation data (for example, 3825). The virtual base location (for example, near rover position 3840) is updated periodically, as indicated by timestamps 4402–4408. Precise satellite data (for example, precise satellite data 3830) are received periodically, as indicated by timestamps 4410–4424. Rover data (for example, rover observations 3825) are received periodically, as indicated by timestamps 4426–4454. The arrivals of the virtual base locations, precise satellite data, and rover data are asynchronous.Each arrival of an epoch of rover data (identified by a corresponding time marker 4426–4454) results in the generation of a corresponding set of virtual base station data (identified by a corresponding time marker 4456–4484). In each case, it is preferred to use the latest virtual base station location, as well as the latest precise satellite data, when processing an epoch of rover observation data with the corresponding virtual base station data. Each pair of rover data and virtual base station data, for example, time marker pair 4424 and 4456, results in the generation of a corresponding rover position, for example, time marker 4485. Generated rover positions are identified by time markers 4485–4499.

[0321] In some embodiments, a new SBS data epoch is created at every time a new rover data epoch is observed. Fig.Figure 45 shows a time graph of a high-precision version of the processing of Fig. 38, Fig. 42 or Fig.43. In this variant, the arrival of a set of precise satellite data (for example, 3830) is used as a trigger for the generation of synthesized base station data (for example, 3850). For example, a set of synthesized base station data (for example, 3850) is generated for each set of precise satellite data (for example, 3850). The virtual base location 3840 (for example, approximate rover position) is updated from time to time, as indicated by timestamps 4502–4508. Precise satellite data (for example, precise satellite data 3830) are received from time to time, as indicated by timestamps 4510–4524. Synthesized base station data (for example, 3850) are generated, for example, by each new set of precise satellite data, as indicated by timestamps 4526–4540.Rover data (for example, Rover observations 3825) are received periodically, as indicated by timestamps 4542–4570. The arrivals of virtual base stations, the precise satellite data, and the rover data are asynchronous, but in this variant, the synthesized base station datasets are synchronized (have the same time tags) as precise satellite datasets, for example, indicated by timestamps 4510 and 4526, 4512 and 4528, etc. Each new epoch of rover data is processed using the most recent synthesized base station dataset. For example, the rover data epochs arriving at timestamps 4544 and 4536 are processed using synthesized base station data prepared at timestamp 4526, etc.

[0322] In some embodiments, a new SBS data epoch is created each time a new precise satellite data set is received. Fig.Figure 46 shows a variant 4600 of the process of Fig. 38, Fig. 42 or Fig.43, in which the virtual base location 4605 is taken from any one of a variety of sources. Some embodiments take as the virtual base location 4605 (a) the autonomous position 4610 of the rover, as determined, for example, by the rover's navigation engine 3845 using rover data 3825. Some embodiments take as the virtual base location 4605 (b) a previous precise rover position 4615, for example, a precise rover position 4220, determined for a previous epoch by differential processor 3855 or by PPP engine 4210. Some embodiments take as the virtual base location 4605 (c) the autonomous position 4610 of the rover, as determined, for example, by the SBS module 3835 using rover data 3825.Some embodiments take as the virtual base station location 4605(d) the autonomous position 4610 of the rover, as determined, for example, by the SBS module 3835 using rover data 3825 and precise satellite data 3830. Some embodiments take as the virtual base station location 4605 an approximate rover location obtained from one or more alternative position sources 4620, such as a rover position determined by an inertial navigation system (INS) 4625 that is colocated to the rover, the position of a mobile phone (cellular phone) tower 4630 near a rover that is colocated to a mobile phone communicating with the tower, user input 4635, such as a location manually entered by a user, for example, with the aid of a keypad 3755 or other user input device, or any desired source 4640 of a virtual base station location.

[0323] Regardless of the source, some embodiments update the virtual base station location 4605 or 3840 from time to time for use by SBS module 3835, as indicated by arrow 4645. For example, the virtual base station location 4605 may be updated: (a) never, (b) for each epoch of rover data, (c) for each nth epoch of rover data, (d) after a predetermined time interval, (e) if the distance between the approximate rover antenna position (for example 3840) or autonomous rover antenna position (for example 4610) and the virtual base station location (for example 4605) exceeds a predetermined threshold, (f) if the distance between the approximate rover antenna position (for example 3840) or autonomous rover antenna position (for example 4610) and the precise rover antenna position (for example 4220) exceeds a predetermined threshold, (g) for each update of the approximate rover antenna position (for example 3840), (h) for each update of the precise rover antenna position (for example, 4220).

[0324] In case (a), the first virtual base station location (e.g., 4605) provided to the SBS module 3835 is used for the entire period during which data processing is performed. In case (b), the virtual base station location (e.g., 4605) is updated each time a new epoch of rover data 3825 is collected, as this new epoch can be used to update the rover approximate position 3840, which can be used as the virtual base station location 4805. In cases (b) and (c), the virtual base station location 4605 is updated each time a certain number (e.g., 1 to 10) of epochs of rover data 3825 are collected. In case (d), the virtual base station location 4605 is updated at a certain time interval (e.g., every 10 seconds).Case (e) can be viewed as a mixture of cases (a) and (b), where the current virtual base station location 4605 is maintained as long as the distance between the current virtual base station location and the approximate rover antenna position is less than a limiting distance (for example, 100 m). Case (f) is similar to case (e), except that the distance between the virtual base station location and a recent precise rover position is used. For case (g), the virtual base station location is updated every time the approximate rover antenna position changes. For case (h), the virtual base station location is updated every time the precise rover antenna position changes.

[0325] In some embodiments, the virtual base station location 3840, which is used to generate the SBS data, comes from the autonomous position solution of the rover receiver, for example, approximate rover position 3840. In some embodiments, the virtual base station location 3840 is not the same position as the autonomous position solution, but somewhere close to it.Some embodiments use as the virtual base station location 3840 a source such as: an autonomously determined position of the rover antenna, a previously determined position of the rover antennas, a synthesized base station data generating module (for example 3835), a precise rover position (for example 4220), a position determined by a PPP engine (for example 4210), a position determined by an inertial navigation system (for example 4625), a mobile phone tower location (for example 3630), information provided by a user (for example 4635), or any other selected source (for example 4540).

[0326] In some embodiments, the virtual base station location 3840 is not kept constant over the rover observation period, but it is updated if certain conditions are met, such as: never, for each epoch of rover data, if the distance between an approximate rover antenna position and the virtual base station location exceeds a predetermined threshold, for each update of the rover antenna position, and for a specific GNSS time interval. In some embodiments, a change in the virtual base station location 3840 is used to trigger the generation of a new SBS epoch of data.

[0327] In some embodiments, the SBS data is used for some type of inter-station differential GNSS processor, regardless of the type of data modeling involved, such as processors using: Auxiliary Inertial Navigation (INS), integrated INS and GNSS processing, normal real-time kinematics (RTK), instant RTK (IRTK, for example, using L1 / L2 for fast instant ambiguity resolution), differential GPS (DGPS) sliding solution processing, and / or triple-differentiated processing. In some embodiments, the SBS data is used in post-processing of the rover data. In some embodiments, SBS data is used in real time (i.e., as soon as the rover observation is available and an SBS record can be generated for it). In some embodiments, the rover's time tag coincides with the SBS data's time tag within a few milliseconds. Part 11.5 SBS References

[0328] Some references relevant to Rover processing include: Leandro RF, Santos, MC and Langley RB, (2006a). “UNB Neutral Atmosphere Models: Development and Performance”. Proceedings of ION NTM 2006, Monterey, California, January, 2006. Leandro RF, Santos, MC and Langley RB, (2006b). “Wide Area Neutral Atmosphere Models for GNSS Applications”. Proceedings of ION GNSS 2006, Fort Worth, Texas, September, 2006.

[0329] Leandro, R.F. (2009). Precise Point Positioning with GPS: A New Approach for Positioning, Atmospheric Studies, and Signal Analysis. PhD dissertation, Department of Geodesy and Geomatics Engineering, Technical Report No. 267, University of New Brunswick, Fredericton, New Brunswick, Canada, page 232. Part 12: Rover Processing with Ambiguity Setting Part 12.1 Ambiguity Setting Introduction

[0330] The conventional high-accuracy absolute positioning solution (also called Precise Point Positioning or PPP) uses precise satellite orbit and time-signal error information. This solution also relies on iono-free observations because no information is available about the ionosphere's behavior at the rover receiver's location (a few centimeters). High-accuracy absolute positioning solutions in this scenario have always relied on floating-point estimations of the carrier phase ambiguities, as it was not possible to maintain the integer nature of these parameters using undifferentiated iono-free observations. Another problem regarding the integer nature of the undifferentiated ambiguities is the presence of non-integer phase biases in the measurements. These biases must also be included in the correction data (e.g.,(time signals) are present, but on the other hand, the ambiguities obtained from the positioning filter will not be of an integer nature.

[0331] Typical observational models used in state-of-the-art processing include: ϕ=R+dT−dt+T+Nif P=R+dT−dt+T where ϕ The phase observation of the rover is based on a satellite signal (measured data), P the code observation is from a satellite signal (measured data) on the rover, R is the distance from satellite to rover during the transmission time of the observed signals. dT is the receiver time signal (also called a code-adapted receiver time signal or receiver code time signal or standard receiver time signal), dt the satellite time signal is (also referred to herein as a code-adapted satellite time signal or satellite code time signal or standard satellite time signal), T is the tropospheric delay of the satellite to the rover signal path, and N if The iono-free ambiguity is.

[0332] A typical state-of-the-art PPP processing uses the ionospheric-free phase ϕ and code P observations (measurements) of signals from multiple satellites together with externally provided satellite time signal information dt in an attempt to calculate values ​​for the receiver position (X). r , Y r , and Z r ) to estimate, as well as receiver time dT, troposphere delay T and the iono-free glide ambiguities NifFloat. The state vector of parameters to be estimated in a Kalman filter implementation is therefore, for each satellite j observed by the rover: {XrYrZrdTTNifjFloat}

[0333] Some embodiments of the invention provide absolute positioning that takes into account the integer nature of carrier-frequency ambiguities in real-time processing at the rover. Real-time processing means that the observation data are processed as soon as: (a) the data are collected; and (b) the necessary information (for example, satellite corrections) is available to do so. Novel procedures use special satellite time signal information so that carrier-phase (also called phase) ambiguity integrity is preserved in the ambiguity state values ​​that are computed at the rover. The rover's processing engine handles the satellite time signal error and a combination of satellite biases that are applied to the receiver observations.

[0334] Prior art PPP engines are unable to use phase-matched time signals or, at least, unable to take advantage of the integer nature of these time signals. Some embodiments of the invention utilize this new information, for example, a modified position engine on rovers.

[0335] One goal of using phase-matched satellite time signals and bias information is to obtain assumed integer ambiguities, which are used to obtain an improved position solution that takes advantage of the integer nature of the ambiguities. This improves position solution (4710) compared to position solution (4705) where the ambiguities are treated as floating-point numbers, as in Fig. 47 shown.

[0336] The result of such a PPP processing of the state of the art does not permit the integer nature of iono-free ambiguities N if to reliably determine, but instead this only allows an estimate of iono-free floating ambiguities. NifFloat. The iono-free glide ambiguities NifFloat can be considered as containing additional effects that can be interpreted as an error e, such that: NifFloat=[αNWLInteger+βNIInteger]+e NifFloat=NifPL+e where an iono-free floating ambiguity is a combination of an iono-free integer ambiguity NifInteger and represents an error e a Widelane integer ambiguity is α is a Widelane ambiguity coefficient. an L1 integer ambiguity is β is an L1 ambiguity coefficient a phase-adapted iono-free ambiguity is

[0337] In some specific cases, it is possible to NifFloat≅Nifiteger to consider in practice. Therefore, there are cases in which the formulation is ambiguous due to the sliding effect. NifFloat as a composition of two other integer ambiguities (as in NifFloat=[αNWLInteger+βNIInteger]+e ) is not necessary.

[0338] Part 12.2 Determining iono-free phase-matched ambiguities NifPL: Option 1

[0339] Some embodiments of the invention are based on eliminating the unwanted effects, or in other words, the iono-free sliding ambiguity error e, to allow the determination of iono-free phase-matched ambiguities. NifPL. To eliminate the error e, the phase observation model is redefined: ϕ=R+dTp−dtp+T+NifPL where dT p a phase-matched receiver time signal (also called a receiver phase time signal), dt p a phase-matched satellite time signal (also called a satellite phase-time signal), and a phase-adapted iono-free ambiguity.

[0340] A satellite phase-time signal dt p In (96) is phase-based and has an integer nature and, unlike equation (94), can provide phase-matched ambiguities free from an error e such that: NifPL=[αNWLInteger+βIInteger]

[0341] The phase-adapted iono-free ambiguity NifPL is not an integer, but has an integer nature; the phase-matched iono-free ambiguity NifPL It can be interpreted as a combination of two integer ambiguities. One way to do this is to assume it is a combination of a Widelane integer ambiguity. NWLInteger and an L1 integer ambiguity NIInteger in which the Widelane ambiguity coefficient α and the L1 ambiguity coefficient are not necessarily integers.

[0342] The phase-matched time signal dt p can be quite different from a standard (code-adjusted) satellite time signal dt. To avoid confusion, the expression dt p used herein to denote the phase-matched satellite time signal, and the expression dt c is used herein to denote the standard (code-adapted) satellite time signal dt c = dt.

[0343] The introduction of the phase-matched satellite time signal expression dtp in (96) means that the corresponding receiver time signal expression dT p is also phase-adapted. The expression dT p is used here to denote the phase-matched receiver time signal and the expression dT c is used to denote the standard (code-adapted) receiver time signal, i.e., dT c = dT. In a positioning engine where phase and code measurements are used simultaneously, two time signal states for the receiver are therefore formed by a model; in this formulation, these are the phase-matched receiver time signal dT. p and the standard (code-adapted) receiver time signal dT c .

[0344] As mentioned above, a typical state-of-the-art PPP processing attempts to estimate values ​​for four parameters for each observed satellite: Receiver coordinates X r , Y r , and Z rReceiver time signal dT, tropospheric delay T and the iono-free floating ambiguities NifFloat. The introduction of phase-matched time signal expressions adds another parameter to be estimated in the rover engine: phase-matched receiver time signal dT p The newly formulated observational models are therefore as shown in equations (91) and (92) and are reproduced here: ϕ=R+dTp−dtp+T+NifPL P=R+dTe−dte+T

[0345] In this formulation, each observation type (phase ϕ or code P) is corrected with its own time signal type (phase-adjusted satellite time signal dt). p or code-adapted satellite time signal dt cSome embodiments of the invention therefore use phase ϕ and code P observations (measurements) of the signals from multiple satellites observed by a rover together with externally provided code-adapted (standard) satellite time signal information dt. c and phase-adapted satellite time signal information dt p for estimating values ​​for distance R, code-adapted receiver time signals dT c , phase-matched receiver time signal dT p , tropospheric delay T and the iono-free phase-matched ambiguities NifPL. The state vector of parameters to be estimated in a Kalman filter implementation is therefore, for each satellite j observed by the rover: {XrYrZrdTcTNifjPL}

[0346] Part 12.3 Determining iono-free phase-matched ambiguities NifPL: Option 2

[0347] A second formulation for handling the phase-adapted information is, for the code-adapted receiver time signal dT c an offset δdT p to substitute what represents the difference between the code-adapted receiver time signal dT c and the phase-matched receiver time signal dT p : δdTp=dTc−dTp so that equations (91) and (92) become accordingly: ϕ=R+dTp−dtp+T+NifPL P=R+dTp+δdtp−dtc+T

[0348] Accordingly, some embodiments of the invention use phase ϕ and code P observations (measurements) of signals from multiple satellites, observed at a rover together with externally provided code-adapted (standard) satellite time signal information dt. c and phase-adapted satellite time signal information dt c for estimating values ​​for distance R, phase-matched receiver time signal dTp , Receiver time signal offset δdT p , tropospheric delay T and the ion-free phase-matched ambiguities NifPL. The state vector of parameters to be estimated in a Kalman filter implementation is therefore, for each satellite j observed by the rover: {XrYrZrdTpδdTpTNifjPL}

[0349] This second formulation still has five types of parameters to be estimated, but because δdT p Since there is an offset, there is less process noise in the Kalman filter than for state vector (100). The advantage of such a model, compared to Option 1, is that the stochastic model can be set up differently, meaning that the noise level associated with a time-signal bias state (e.g., δdT) can be adjusted. p ) is assigned, may differ from a time signal state (e.g. dT). cAssuming that phase- and code-matched time signals behave in a similar way, the noise level required to model a bias state should be lower than for modeling a time signal state.

[0350] Part 12.4 Determining iono-free phase-matched ambiguities {XrYrZrdTpδdTpTNifjPL} Option 3

[0351] A third formulation for handling the phase-matched information is to use the phase-matched receiver time signal dT. p an offset -δdT p to substitute, which is the difference between the phase-matched receiver time signal dT p and the code-adapted receiver time signal dT c represents: −δdTp=dTp−dTc so that equations (91) and (92) become: ϕ=R−dTp+dTc−dtp+T+NifPL P=R+dTc+δdTc−dtc+T

[0352] Accordingly, some embodiments of the invention use phase ϕ and code P observations or observations (measurements) of the signals from multiple satellites, observed at a rover, together with externally provided code-adapted (standard) satellite time signal information dt. c and phase-adapted satellite time signal information dt p for estimating values ​​for distance R, code-adapted receiver time signals dT c , Receiver time signal offset -δdT p , tropospheric delay T and the iono-free phase-matched ambiguities NifPL. The state vector of the parameters to be estimated in a Kalman filter implementation is therefore, for each satellite j observed by the rover: {XrYrZrdTcδdTpTNifjPL}

[0353] This third formulation still has five types of parameters to be estimated, but because - δdT pWith an offset, there is less process noise in the Kalman filter than for state vector (100).

[0354] Part 12.5 Determining iono-free phase-matched ambiguities NifPL Option 4

[0355] A fourth formulation for handling phase-adapted information first estimates the ion-free floating ambiguities. NifjFloat using code-adapted (standard) satellite time signals dt c for phase observations ϕ and for code observations P as in (95), and shifts the iono-free floating ambiguities NifjFloat then using the phase-matched time signal information to obtain the iono-free phase-matched ambiguities NifPL.

[0356] The starting point for this formulation is: ϕ=R+dTc−dtc+T+NifFloat P=R+dTc−dtc+T

[0357] It should be noted that (109) and (110) are identical to (91) and (92), since dT c = dT and dt c = dt .

[0358] This fourth formulation, like the typical state-of-the-art PPP processing discussed above, uses phase ϕ and code P observations (measurements) of signals from multiple satellites, observed at a rover, together with externally provided code-adapted (standard) satellite time signal information dt. c for estimating values ​​for distance R, code-adapted receiver time signal dT c , tropospheric delay T and the iono-free glide ambiguities NifFloat. The state vector of the parameters to be estimated in a Kalman filter implementation is therefore, for each satellite j observed by the rover: {XrYrZrdTcδdTpTNifjFloat}

[0359] As with the PPP processing of the state of the art, the estimation of parameter values ​​of the state vector (111) does not allow the iono-free floating ambiguities NifjFloat, to be reliably determined as iono-free phase-matched ambiguities NifPL: NifFloat=NifPL+e=[αNWLInteger+βNIInteger]+e where an iono-free floating ambiguity is a combination of an iono-free phase-matched ambiguity NifPL and represents an error e, a Widelane integer ambiguity is, α is a Widelane ambiguity coefficient, an L1 integer ambiguity, and β is an L1 ambiguity coefficient.

[0360] This fourth formulation assumes that error e represents the difference between the code-adapted (standard) satellite time signal dt cand the phase-matched satellite time signal dt p : e=dtc−dtp so that NifPL=NifFloat−(dte−dtp) NifPL=[αNWLInteger+βNIInteger]+e−e NifPL=[αNWLInteger+βNIInteger]

[0361] In summary, this means that this fourth formulation retains iono-free glide ambiguities. NifFloat from (109) and then shifts each of these by the difference between the standard (code-adapted) satellite time signal dtcj and the phase-matched time signal dtpj for the corresponding satellite j to obtain the iono-free phase-matched ambiguities NifPL as shown in (114).

[0362] For this purpose, the rover can be provided with (1) the standard (code-adapted) satellite time signal. dtcj and the phase-matched time signal dtpj for each satellite j, observed at the rover, or (2) the standard (code-adapted) satellite time signal dtcj and a time signal bias δt c , which represents the difference between the standard (code-adapted) satellite time signal dtcj and the phase-matched time signal dtpj, or (3) the phase-matched time signal dtpj and a time signal bias δt p , which represents the difference between the phase-matched time signal dtpj and the standard (code-adapted) satellite time signal dtcj. These are equivalent for processing purposes, as can be seen from (114).

[0363] This fourth formulation has advantages and disadvantages. One disadvantage is that it assumes the behavior of the standard (code-adapted) satellite time signal. dtcj of the phase-matched time signal dtpj For each satellite j observed by the rover, essentially the same applies over the period of time for which the ambiguity was calculated. One advantage is that jumps (integer cycle shifts) that can occur in the phase-time signal estimates can be more easily handled in processing on the rover.

[0364] Part 12.6 Determining Position Using Melbourne-Wübbena Biases bWLj:

[0365] After the iono-free phase-matched ambiguities NifjPL Destined for a given era, they can be differentiated individually as follows: ∇abNifPL, Individually differentiated integer (fixed) Widelane ambiguities ∇abNWLInteger can be removed from these to obtain individually differentiated L1 glide ambiguities ∇abNIFloat, and the individually differentiated L1 gliding ambiguities ∇abNIFloat, can be defined as individually differentiated Li integer ambiguities ∇abNIInteger.

[0366] Individually differentiated Widelane gene count ambiguities ∇abNWLInteger The values ​​are estimated in a Widelane ambiguity filter that runs in parallel to the geometric filter of the rover's processing engine. The rover receiver is provided with a Melbourne-Wübbena bias. bMWj for each of the j satellites in sight, for example for an external source of data corrections - see Part 7 (MW bias processor). Melbourne-Wübbena bias bMWj can be calculated as: ϕWLj−PNLjλWL=NWLjInteger+bMWj−bMWR where the Widelane carrier phase combination of rover observations from satellite j is the narrowlane code combination of rover observations from satellite j is λ WL the Widelane wavelength is the Widelane integer ambiguity for satellite j is the Melbourne-Wübbena bias for the R is.

[0367] The Widelane ambiguity filter eliminates the Melbourne-Wübbena bias. bMLR of the rover by differentiating (115) for each pair of satellites “a” and “b” to obtain the individually differentiated Widelane integer ambiguities ∇abNWLInteger: ∇abNWLInteger=∇ab(ϕWL−PNL¯λWL)−∇abbMW where ∇abbMW=bMWa−bMWb

[0368] As soon as the individually differentiated Widelane integer ambiguities ∇abNWLInteger Once known, they are removed from the individually differentiated iono-free phase-matched ambiguities. ∇abNifPL to preserve the individually differentiated L1 gliding ambiguities ∇abNIFloat.

[0369] The float Kalman filter, used to estimate iono-free phase-matched ambiguities NifPL (or the iono-free glide ambiguities) NifjFloat (in the third alternative formulation, discussed above) there is also C Nij , the covariance matrix of the iono-free phase-matched ambiguities NifPL. Because the Widelane ambiguities are integer (fixed) values, once their values ​​are found, C Nid the same for NifPL as for NifjFloat.

[0370] Of (95) it is known that: ∇NifabPL=α∇NWLabInteger+β∇NIabFloat where the individually differentiated phase-adapted ion-free ambiguity is (differentiated between satellite “a” and “b”) the individually differentiated Widelane integer ambiguity is (differentiated between satellites “a” and “b”) the individually differentiated L1 glide ambiguity is (differentiated between satellites “a” and “b”)

[0371] Therefore, the following applies ∇NIabFloat=∇NifabPL−α∇NWLabintegerβ

[0372] Because ∇NWLabInteger Since fixed (integer) values ​​are assumed, the corresponding covariant matrices or covariance matrices of the L1 floating ambiguities are assumed. NIabFloat and the iono-free phase-matched ambiguities NifjPL related by: CNIFloat=CNifPL⋅F2 where the covariant matrix of L1 floating ambiguities NIjFloat is C NifPL the covariant matrix of iono-free phase-matched ambiguities NifjPL is F is a factor used to convert the units of the variances from iono-free cycles to L1 cycles.

[0373] The desired “fixed” (integer nature) individually differentiated L1 gliding ambiguities ∇NIabInteger can be determined by the individually differentiated L1 gliding ambiguities ∇NIabFloat and the covariant matrix CNIFloat the L1 glide ambiguities NIabFloat using well-known techniques such as Lambda (Jonge et al. 1994), modified Lambda (Chan et al. 2005), weighted averaging of candidate sets, or others.

[0374] After determining the individually differentiated integer Widelane ambiguities ∇NWLabInteger and the individually differentiated integer L1 ambiguities, iono-free ambiguities are ∇NifabInteger of an integer nature determined by: ∇NifabInteger=α∇NWLabInteger+β∇NIabInteger

[0375] The iono-free ambiguities of an integer nature ∇NifabInteger are introduced (“pressed in”) as a pseudo-observation into the Kalman glide filter (or optionally a copy of it) to determine a rover position based on integer ambiguities. The state vector of the glide filter copy is therefore: {XrYrZrdTcT}

[0376] The rover position can then be determined with an accuracy (and precision) that is essentially better than for typical prior art PPP processing, where ambiguities are treated as floating-point numbers, as in Fig. 47 shown.

[0377] Fig.Figure 48 shows a block diagram of a positioning scheme 4800 using an ambiguity setting according to some embodiments of the invention. In this example, after satellite time signal (time signal and orbit) corrections 4802 and MW bias corrections 4804 are received, for example via a link such as communication satellite 4806, they are used together with GNSS data 4808, collected by the rover receiver 4810, for example from observations of GNSS satellites 4812, 4814, 4816. Processing is carried out in two filters: a first filter 4818 with a geometric ionospheric-free model (Part 12.5, Equations 107-108) and a second filter 4820 with a geometry- and ionospheric-free model (Part 12.6, Equation 116). From the first filter 4818, phase-matched ionosphere-free ambiguities 4822 with corresponding covariant matrix 4824 are obtained.The second filter, 4820, yields integer Widelane ambiguities 4826. At 4828, single-differentiated glide ambiguities of integer nature 4830 are determined from the phase-matched ionosphere-free ambiguities 4822 and the Widelane ambiguities 4826. The single-differentiated glide ambiguities of integer nature 4830 are processed again at 4832 using an integer least squares-based approach (for "setting" these, since this expression is defined herein as meaning setting them as integers or weighted averages) to obtain single-differentiated L1 ambiguities 4834 of integer nature. These ambiguities 4834 are then recombined at 4836 with the Widelane ambiguities of integer nature 4826 to obtain single-differentiated ionosphere-free ambiguities of integer nature 4840.These ambiguities 4840 are then applied in a third filter 4842, which estimates a rover-receiver time signal 4844 and satellite-rover distances 4846, from which the receiver position is determined.

[0378] Fig.Figure 49 shows a block diagram of a positioning scheme 4900 using an ambiguity determination according to some embodiments of the invention. In this example, satellite time signal corrections 4902, received, for example, via a link such as communication satellites 4904, are used together with GNSS data 4906, collected at the rover receiver 4908, for example, from observations of GNSS satellites 4910, 4912, 4814, in a first filter 4916. Filter 4916 estimates floating-point values ​​for carrier ambiguities 4918 with corresponding covariances 4920. At 4922, carrier phase ambiguities of an integer nature 4924 are determined using floating ambiguities 4918 and the corresponding covariant matrix 4920, provided by a first filter 4916.The carrier phase ambiguities of an integer nature 4924 and satellite time signal corrections 4902 and GNSS data 4906 are used in a second filter 4926 to determine the receiver position 4928.

[0379] Fig.Figure 50 shows a block diagram of a positioning scheme 5000 using an ambiguity determination according to some embodiments of the invention. In this example, satellite time signal corrections 5002, received, for example, via a link such as a communications satellite 5004, are used together with GNSS data 5006, collected at the rover receiver 5008, for example, from observations of GNSS satellites 5010, 5012, 5014, in a first filter 5016. Filter 5016 estimates values ​​for carrier ambiguities 5018 with corresponding covariances 5020. At 5022, individually differentiated carrier phase ambiguities of an integer nature 5024 are determined using floating ambiguities 5018 and the corresponding covariant matrix 5020, provided by the first filter 5016.The individually differentiated carrier-phase ambiguities of an integer nature 5024 and satellite time signal corrections 5002 and GNSS data 5006 are used in a second filter 5026 to determine the receiver position 5028.

[0380] Fig.Figure 51 shows a block diagram of a positioning scheme 5100 using an ambiguity setting according to some embodiments of the invention. In this example, satellite time signal corrections 5102 and satellite MW biases 5104, received, for example, via a link such as a communications satellite 5106, are used together with GNSS data 5108, collected at the rover receiver 5110, for example, from observations of GNSS satellites 5112, 5114, 5116, in a first filter 5118. The filter 5118 estimates values ​​for carrier-phase ionospheric-free glide ambiguities 5120 with corresponding covariances 5122. A third filter 5124 uses the satellite MW bias corrections 5104 and the GNSS data 5108 to determine widelane carrier ambiguities 5126.At 5128, carrier phase ambiguities of an integer nature 5130 are determined using the floating-iono-free ambiguities 5120 and the corresponding covariant matrix 5122, provided by the first filter 5118, and the wide-lane carrier ambiguities 5126, provided by the third filter 5124. A second filter 5132 uses the carrier phase ambiguities of an integer nature 5130 with the GNSS data 5108 and the satellite time signal corrections 5102 to determine a receiver position 5134.

[0381] Fig. Figure 52 shows a block diagram of a positioning scheme 5200 using an ambiguity setting according to some embodiments of the invention. In this variant of the example of Fig.Satellite time signal corrections 5202 and satellite MW biases 5204 are received, for example, via a link such as a communications satellite 5206. Satellite time signal corrections 5202 are used together with GNSS data 5208, collected by the rover receiver 5210, for example, from observations of GNSS satellites 5212, 5214, 5216, in a first filter 5218. The first filter 5218 estimates values ​​for glide ambiguities 5220 with corresponding covariances 5222. A third filter 5224 uses the satellite MW bias corrections 5204 and the GNSS data 5208 to determine widelane carrier ambiguities 5228. In 5228, integer ambiguities 5230 are determined by the carrier ambiguities 5220, the carrier ambiguities covariances 5222 and the Widelane carrier ambiguities 5226, using a covariant matrix that is scaled by a certain factor (see, for example, Equation 120).A second filter 5232 uses carrier-phase ambiguities of an integer nature 5230 with the GNSS data 5208 and the satellite time signal corrections 5202 to determine a receiver position 5234.

[0382] Fig. Figure 53 shows a block diagram of a positioning scheme 5300 using an ambiguity setting according to some embodiments of the invention. In this variant of the examples of Fig. 51 and Fig.Satellite time signal corrections 5302 and satellite MW biases 5304 are received, for example, via a link such as a communications satellite 5306. Satellite time signal corrections 5302 are used together with GNSS data 5308, collected by the rover receiver 5310, for example, from observations of GNSS satellites 5312, 5314, 5316, in a first filter 5318. The first filter 5318 estimates values ​​for glide ambiguities 5320 with corresponding covariances 5322. A third filter 5324 uses the satellite MW bias corrections 5304 and the GNSS data 5308 to determine single-differentiated inter-satellite wide-lane carrier ambiguities 5326.In 5328, single-differentiated inter-satellite carrier phase ambiguities of an integer nature 5330 are determined from the carrier phase ambiguities 5320, the carrier ambiguity covariances 5322, and the wide-lane carrier ambiguities 5326, using a covariant matrix scaled by a certain factor (see, for example, equation 120). A second filter 5232 uses the carrier phase ambiguities of an integer nature 5330 with the GNSS data 5308 and the satellite time signal corrections 5302 to determine a receiver position 5334.

[0383] Fig.Figure 54 shows a block diagram of a positioning scheme 5400 using an ambiguity setting according to some embodiments of the invention. In this variant, the satellite time signal corrections 5402 and satellite MW biases 5404 are received, for example, via a link such as a communications satellite 5406. Satellite time signal corrections 5402 are used together with GNSS data 5408, collected by the rover receiver 5410, for example, from observations of GNSS satellites 5412, 5414, and 5416, in a first filter 5418. The first filter 5418 estimates values ​​for glide ambiguities 5420 with corresponding covariances 5422. A third filter 5424 uses the satellite MW bias corrections 5404 and the GNSS data 5408 to determine widelane carrier ambiguities 5426.In 5428, iono-free integer ambiguities 5430 of the two carrier frequencies (for example, L1 and L2) are determined by the carrier ambiguities 5420, the carrier ambiguity covariances 5422, and the Widelane carrier ambiguities 5426, using a covariant matrix scaled by a certain factor (see, for example, Equation 120). A second filter 5432 uses the integer carrier phase ambiguities 5430 with the GNSS data 5408 and the satellite time signal corrections 5402 to determine a receiver position 5434. The integer carrier ambiguities 5430 can be sets of L1 and L2 ambiguities, sets of L2 and L5 ambiguities, or sets of any linear combination of two or more GNSS frequencies.

[0384] Fig.Figure 55 shows a block diagram of a positioning scheme 5500 using an ambiguity setting according to some embodiments of the invention. In this variant, satellite time signal corrections 5502, ionospheric delay information 5504, and satellite MW biases 5506 are received, for example, via a link such as a communications satellite 5508. Satellite time signal corrections 5502 and ionospheric delay information 5504 are used together with GNSS data 5510, collected by the rover receiver 5512, for example, from observations of GNSS satellites 5514, 5516, and 5518, in a first filter 5520. The first filter 5520 estimates values ​​for glide ambiguities 5522 with corresponding covariances 5524. A third filter 5526 uses the satellite MW bias corrections 5506 and the GNSS data 5510 to determine Widelane carrier ambiguities 5528.In 5530, L1 ambiguities of an integer nature 5532 are determined from the carrier ambiguities 5522, the carrier ambiguity covariances 5524, and the Widelane carrier ambiguities 5528 using a covariant matrix scaled by a certain factor (see, for example, Equation 120). A second filter 5534 uses the L1 carrier phase ambiguities of an integer nature 5532 with the GNSS data 5510, the satellite time signal corrections 5502, and ionospheric delay information 5504 to determine the receiver position 5536. In further variants of . Fig. 55 The ionospheric delay information 5504 is used to feed to one or more filters, used in data processing, and / or the specific ambiguities of an integer nature 5532, may be sets of L1 ambiguities or L2 ambiguities or L5 ambiguities or sets of ambiguities of any GNSS frequency.

[0385] Fig. 56 is the same as Fig. 54, in which the second filter 5632 is an independent filter from the first filter 5418 and the third filter 5424.

[0386] Fig. Figure 57 shows a variation 5700 of the embodiment of Fig. 56, in which the second filter 5732 is a copy of the first filter 5418.

[0387] Fig. Figure 58 shows a variation 5800 of the embodiments of Fig. 56 and Fig. 57, in which the second filter 5632 of Fig. 56 and Fig. 5732 of Fig. 57 is replaced by the first filter 5418. This means that when the ionosphere-free ambiguities of an integer nature 5430 are determined, they are returned to the first filter 5418, which is then used to determine the receiver position 5434.

[0388] Fig.Figure 59 shows a detailed view of an embodiment of the first filter 5418, and shows that the satellite time signal corrections 5402 include code-matched satellite time signal corrections 5906 and phase-matched satellite time signal corrections 5908. The code-matched satellite time signal corrections 5906 are used at 5914 to model GNSS observations in an observation filter 5912. The resulting code-matched floating carrier ambiguities 5918 are adjusted at 5920 to the level of the phase-matched satellite time signals 5908 by applying the difference between the code-matched satellite time signals 5906 and the phase-matched satellite time signals 5908 to obtain phase-matched carrier ambiguities 5420. Observation filter 5912 also provides covariances 5422 of the carrier ambiguities.

[0389] Fig.Figure 60 shows a detailed view of an embodiment of the first filter 5418, in which the code-matched satellite time signal corrections 5906 are used in an observation filter 6012 for modeling code observations at 6014. The phase-matched satellite time signal 5908 is used at 6016 for modeling carrier phase observations. The resulting ambiguities from the observation filter 6012 are the phase-matched carrier ambiguities 5420. The observation filter 6012 also provides covariances 5422 of the carrier ambiguities.

[0390] Fig.Figure 61 shows a variant of previous figures in which the integer ambiguities are determined at 6128 using at least one of: rounding the floating ambiguity to the nearest integer, choosing the best integer candidates from a set of integer candidates generated using integer least squares, and calculating floating values ​​using a set of integer candidates generated using integer least squares.

[0391] Fig.Figure 62 shows a variant of previous figures in which at least one of the first filter 5418 and the second filter 5432 further estimates values ​​6202, 6252 including at least one of the values ​​for: receiver phase-matched time signal states 6204, 6254, receiver code-matched time signal states 6606, 6256, tropospheric delay states 6208, 6258, receiver time signal bias states 6210, 6260, representing a difference per satellite between code-matched receiver time signal and phase-matched receiver time signal and multipath states 6212, 6262. Part 12.7 Ambiguity-Defining References

[0392] Some references relevant for determining ambiguity include: Jonge de, PJ, and CCJM Tiberius (1994). A new GPS ambiguity estimation method based on integer least squares. Proceedings of the 3rd International Symposium on Differential Satellite Navigation Systems DSNS'94, London, UK, April 18-22, Journal No. 73, pages 9ff. X.-W. Chang, X. Yang and T. Zhou (2005). MLAMBDA: a modified LAMBDA method for integer least-squares estimation. Journal of Geodesy, Springer Berlin / Heidelberg. Volume 79, Number 9 / December, 2005, pages 552-565. Part 13: Summary of some of the inventive concepts Section 13A: MW (Melbourne-Wübbena) bias processing (A-2565) 1. [MW processing] A method for processing a set of GNSS signal data derived from code observations and carrier phase observations at multiple receivers of GNSS signals from multiple satellites over multiple epochs, wherein the GNSS signals have at least two carrier frequencies, comprising: Forming one MW (Melbourne-Wübbena) combination per receiver-satellite pair at each epoch to obtain one MW data set per epoch, and Estimating from the MW datasets per epoch an MW bias per satellite, which may vary from epoch to epoch, and a set of WL (widelane) ambiguities, where each WL ambiguity corresponds to one of a receiver-satellite link and one of a satellite-receiver-satellite link, wherein the MW bias per satellite is represented by a model as one of (i) a single estimation parameter and (ii) an estimated offset plus harmonic variations with estimated amplitudes. 2. The procedure according to 1, further comprising the application of corrections to the GNSS signal data. 3. The method according to one of 1-2, further comprising smoothing at least one linear combination of GNSS signal data before estimating an MW bias per satellite. 4. The procedure according to one of 1-3, furthermore including the application of at least one MW bias condition. 5. The procedure according to one of 1-4, furthermore including the application of at least one integer WL ambiguity condition. 6. The procedure according to one of 1-4, further comprising using a spanning tree (ST) over an observation graph and a filter graph to restrict the WL ambiguities. 7. The procedure according to one of 1-4, further comprising using a minimum spanning tree (MST) on an observation graph and a filter graph to restrict the WL ambiguities. 8. The method according to 7, wherein the minimum spanning tree is based on edge weights, each edge weight being derived from a receiver satellite geometry. 9. The method according to 8, wherein the edge weighting is defined with reference to one of (i) a geometric distance from receiver to satellite, (ii) a satellite elevation angle, and (iii) a geometric distance from one satellite to receiver to satellite, and (iv) a combination of the altitudes at which the two satellites are seen in a single-differentiated combination at a station. 10. The procedure according to 7, wherein the minimum spanning tree is based on edge weights, each edge weight being based on WL ambiguity information, defined with reference to one of (i) a difference from a WL ambiguity to an integer, (ii) WL ambiguity variance and (iii) a combination of (i) and (ii). 11. The procedure according to a number from 1-10, furthermore including defining at least one of the WL ambiguities as an integer value. 12. The procedure according to one of 1-10, further comprising determining candidate sets of WL integer ambiguity values, forming a weighted combination of the candidate sets and specifying at least one of the WL ambiguities as a value to be taken from the weighted combination. 13. The procedure according to one of 11-12, wherein the estimation step includes introducing fixed WL ambiguities so that MW biases are estimated which are compatible with fixed WL ambiguities, 14. The procedure according to 13, wherein the estimation step comprises: applying an iterative filter to the MW data per epoch and wherein introducing fixed WL ambiguities comprises (i) inputting the fixed WL ambiguities as observations into the filter, (ii) inputting the fixed WL ambiguities as observations into a copy of the filter, generated after each of a plurality of observation updates, and (iii) reducing the MW combinations by the fixed WL ambiguities and inputting the resulting reduced MW combinations into a second filter without ambiguity states to estimate at least one MW bias per satellite. 15. The procedure according to one of 1-14, further comprising shifting at least one MW bias by an integer number of WL cycles. 16. The procedure according to one of 1-14, further comprising shifting at least one MW bias and its corresponding WL ambiguity by an integer number of WL cycles. 17. The procedure according to one of 1-16, wherein the navigation message includes orbit information. 18. Device designed to carry out the method according to one of 1-17. 19. A computer program, comprising instructions, configured, when executed on a computer processing unit, to perform a procedure according to one of 1-17. 20. A computer-readable medium comprising a computer program according to 19. Part 13B: Orbit Processing (A-2647) 1. A method for processing a set of GNSS signal data derived from signals from GNSS satellites observed at reference station receivers, wherein the data represent code observations and carrier observations on each of at least two carriers over several epochs, comprising: a. Obtaining an orbit start vector (2635), comprising a time sequence of predicted positions and predicted velocities for each satellite over a first interval and partial derivatives of the predicted positions and predicted velocities with respect to initial positions, initial velocities, force model parameters and Earth orientation parameters, b. Obtaining ionospheric-free linear combinations (2645) of the code observations and the carrier observations for each satellite at several reference stations, and c. iterative correction of the orbit start vector (2635) using at each epoch the ionosphere-free linear combinations (2645) and predicted Earth orientation parameters (2610) as soon as the ionosphere-free linear combinations of the epoch are available, to obtain updated orbit start vector values ​​(2680), comprising a time sequence of predicted positions and predicted velocities for each satellite over a subsequent interval of epochs and an estimate of Earth orientation parameters. 2. [Start] The procedure according to 1, wherein obtaining an orbit start vector (2635) includes: i. Obtaining an approximate orbit launch vector (2615) for the satellites, ii. Obtaining predicted Earth orbit parameters (2610), iii. iterative integration of the approximate orbit vector with the predicted Earth orbit parameters to obtain an orbit prediction (2620) for an initial time interval, and with each iteration, fitting the orbit prediction (2620) to the approximate orbit vector, and iv. Creating from the orbit prediction (2620) an initial set of values ​​for the orbit start vector and partial derivatives (2635). 3. The procedure according to 2, wherein the approximate orbit vector (2615) is obtained from one of: a transmitted satellite navigation message, IGS ultrafast orbit data and other sources of predicted orbits. 4. The procedure according to one of 2-3, wherein an adjustment of the orbit prediction (2620) to the approximate orbit vector (2615) is performed using a least squares approach. 5. The procedure according to one of 2-4, wherein integrating the approximate orbit vector (2615) with the predicted Earth orientation parameters (2610) to obtain an orbit prediction (2620) is iterated until the orbit prediction remains essentially constant. 6. The procedure according to 1, wherein obtaining an orbit start vector (2635) comprises creating the orbit start vector from a set of updated orbit start vector values ​​(2680) that is not older than a predetermined time interval. 7. The procedure according to 6, wherein the predetermined time interval is no more than a few hours. 8. [Operation] The procedure according to one of 6-7, comprising creating the orbit launch vector (2635): Mapping a new orbit start vector (2690) from the updated orbit start vector (2680) and integrating the new orbit start vector (2690) to obtain new values ​​for the orbit start vector (2635). 9. The procedure according to 8, wherein integrating the new orbit start vector (2690) includes integrating the new start vector using Earth orientation parameters from the updated start vector values ​​(2680). 10. [Kalman] The procedure according to one of 1-9, wherein a correction includes applying an iterative filter, comprising one of: a Kalman filter, a UD-factored filter and a square root information filter. 11. [Satellite parameters] The procedure according to one of 1-10, wherein the updated orbit state vector (2680) further includes: additional parameters for each satellite, and wherein correcting the orbit start vector includes correcting the additional parameters for each satellite. 12. [Issue] The procedure according to one of 1-11, further comprising: mapping values ​​from the updated orbit start vector (2680) to a present epoch to obtain a present-epoch orbit position and velocity for each satellite. 13. [Define] The procedure according to one of 1-6, wherein the orbit start vector further comprises an ionospheric-free ambiguity (2575) per receiver-satellite pair, wherein correcting this orbit start vector (2635) comprises estimating floating-point values ​​for the ionospheric-free ambiguities, and wherein the procedure further comprises: 1. Obtaining a value for a Widelane ambiguity (340) per receiver-satellite pair, where the Widelane ambiguity values ​​are of an integer nature, 2. Determining values ​​for ambiguities of an integer nature that are linearly related to the Widelane ambiguities and the ionosphere-free ambiguities of the values ​​of the Widelane ambiguities and the floating-point values ​​of the ionosphere-free ambiguities, 3. Determining the values ​​of the ionosphere-free ambiguities using values ​​of an integer nature, and 4. when the values ​​of the ionospheric-free ambiguities are determined, iterative correction of the orbit start vector (2635) using a time sequence of the ionospheric-free linear combinations (2645) and a set of Earth orbit parameters to obtain an updated orbit start vector (2680), comprising a time sequence of predicted positions and predicted velocities for each satellite over an interval of several epochs and an estimation of the Earth orientation parameters. 14. The procedure according to 13, wherein the ambiguities that are linearly related to the widelane ambiguities and the ionosphere-free ambiguities comprise: narrowlane ambiguities, L1 ambiguities and L2 ambiguities. 15. [Epoch] The procedure according to one of 1-14, where the epochs occur at a rate of approximately 1 Hz. 16. [Filter update] The procedure according to one of 1-15, wherein an iterative correction of the orbit start vector includes an estimation for each epoch of the values ​​of a satellite time signal for each satellite and a satellite position for each satellite at each epoch. 17. [Filter update] The procedure according to one of 1-15, wherein an iterative correction of the orbit start vector includes an estimation for each epoch of the values ​​of a satellite time signal for each satellite, a satellite time signal drift, a satellite time signal drift rate and a satellite position for each satellite at each epoch. 18. [Orbit estimation] The procedure according to one of 1-17, wherein the predicted time sequence of approximate positions for each satellite covers an interval of at least 150 seconds for at least some of the epochs. 19. [Worldwide Network] The procedure according to one of 1-18, wherein the reference stations are widely distributed around the Earth and the GNSS signal data from each reference station represent code observations and carrier observations of a subset of the GNSS satellites at each epoch. 20. Device designed to carry out the method according to one of 1-19. 21. A computer program, comprising instructions, configured, when executed on a computer processing unit, to perform a procedure according to one of 1-19. 22. A computer-readable medium comprising a computer program according to 21. Part 13C: Phase-Time Signal Processing (A2585) 1. [Network Processing - Estimating Phase-Time Signals] A method for processing a set of GNSS signal data derived from code observations and carrier phase observations at multiple receivers of GNSS signals from multiple satellites over multiple epochs, wherein the GNSS signals have at least two carrier frequencies, and a navigation message containing orbit information, comprising: a. Obtaining precise orbit information for each satellite, b. Determining at least one set of ambiguities per receiver, where each ambiguity corresponds to one from a receiver-satellite link and a satellite-receiver-satellite link, and c. Using at least the precise orbit information, ambiguities, and GNSS signal data to estimate a phase-matched time signal per satellite. 2. The procedure according to 1, wherein a determination of ambiguities is performed at a first rate and wherein an estimation of a phase-matched time signal per satellite is performed at a second rate higher than the first rate. 3. The procedure according to one of 1-2, comprising an estimation of the phase-matched time signal per satellite: i. Using at least the precise orbit information, ambiguities, and GNSS signal data to estimate a set of phase-matched time signals per receiver, each phase-matched time signal corresponding to one from a receiver-satellite link and one satellite-receiver-satellite link, and ii. Using a multitude of phase-matched time signals to estimate one phase-matched time signal per satellite. 4. The procedure according to one of 1-3, wherein determining the ambiguities includes using at least one phase-matched time signal per satellite, previously estimated to estimate the ambiguities. 5. The method according to one of 1-4, further comprising obtaining at least one additional phase-matched time signal per satellite, estimated from an external source, and wherein determining the ambiguities comprises using the at least one additional phase-matched time signal per satellite to estimate the ambiguities. 6. The procedure according to one of 1-5, furthermore comprehensive 1. Determining at least one set of ambiguities per receiver for additional receivers, where each ambiguity corresponds to one from a receiver-satellite link and a satellite-receiver-satellite link, 2. after determining the ambiguities for the additional receivers, using at least the precise orbit information, the ambiguities for the additional receivers and the GNSS signal data to estimate at least one additional phase-matched time signal per satellite. 7. The method according to one of 1-6, wherein the at least two carrier frequencies include at least two of the GPS L1, GPS L2 and GPS L5 frequencies. 8. The procedure according to one of 1-7, wherein determining at least one set of ambiguities per receiver includes at least one of: estimating floating ambiguity values, estimating floating ambiguity values ​​and fixing the floating ambiguity values ​​to integer values, estimating floating ambiguity values ​​and forming at least one weighted average of the integer value candidates, and restricting the ambiguity values ​​in a sequential filter. 9. The procedure according to one of 1-8, where the ambiguities are undifferentiated between satellites. 10. The procedure according to one of 1-8, where the ambiguities are differentiated individually between satellites. 11. Device designed to carry out the method according to one of 1-10. 12. A computer program, comprising instructions, configured, when executed on a computer processing unit, to perform a procedure according to one of 1-10. 13. A computer-readable medium comprising a computer program as defined in 12. PART 13D: Rover processing with synthesized reference station data (A2617) [Summary (Fig. 38)] 1. [SBS processing] A method for determining the position of a rover antenna, comprising: a. Obtaining rover GNSS data derived from code observations and carrier phase observations of GNSS signals from multiple satellites over multiple epochs, b. Obtaining precise satellite data for the satellites, c. Determining a virtual base station location, d. Generating epochs of synthesized base station data using at least the precise satellite data and the virtual base station location, and e. Applying a differential process to at least the rover GNSS data and the synthesized base station data to determine at least rover antenna positions. 2. [low latency] The method according to 1, wherein generating epochs of synthesized base station data includes generating an epoch of virtual base station data for each epoch of GNSS rover data. 3. [High Accuracy] The method according to 1, wherein obtaining precise satellite data includes obtaining precise satellite data in sets, wherein generating epochs of synthesized base station data includes generating an epoch of synthesized base station data for each set of precise satellite data, and wherein applying a differential process includes matching each set of GNSS rover data with an epoch of synthesized base station data. 4. [Virtual Base Station Location] The procedure according to one of 1-3, wherein determining a virtual base station location comprises determining the virtual base station location of at least one of: an autonomously determined position of the rover antenna, a pre-determined position of the rover antenna, a synthesized base station data generation module, an inertial navigation system, a mobile phone tower, and information provided by a user. 5. The procedure according to one of 1-4, further comprising an update of the virtual base station location upon the occurrence of at least one of: Never, for any epoch of rover data, if the distance between an approximate rover antenna position and the virtual base station location exceeds a predetermined threshold, and for each update of the rover antenna position. 6. The procedure according to one of 1-5, wherein each epoch of synthesized base station data is generated for a corresponding virtual base station location. 7. The procedure according to one of 1-5, wherein determining the virtual base station location includes selecting a virtual base station location near a present approximate rover antenna position. 8. The procedure according to one of 1-7, wherein a new virtual base position is determined when one or more of the following criteria are met: Each rover epoch, each nth rover epoch, after a time interval, after exceeding a distance between a present approximate rover position and a present virtual base station location. 9. The procedure according to one of 1-8, whereby the virtual base station location is generated for a specific GNSS time interval. 10. The method according to one of 1-9, wherein applying a differential process to at least the rover GNSS data and the synthesized base station data to determine at least rover antenna positions comprises at least one of: auxiliary inertial navigation (integrated inertial navigation and GNSS) processing, real-time kinematics (RTK) processing, instant real-time kinematics (IRTK) processing; differential GPS processing; floating-point processing; triple-differentiated processing; post-processing and real-time processing. 11. The procedure according to one of 1-10, further comprising a matching of each epoch of rover GNSS data with an epoch of synthesized base station data within a few milliseconds. 12. The method according to one of 1-11, wherein generating epochs of synthesized base station data comprises generating a set of synthesized base station observations for each of a plurality of discrete times, and wherein applying a differential process comprises processing each epoch of GNSS rover data with a set of synthesized base station observations for a discrete time that is within ten seconds of the epoch of the GNSS rover data being processed. 13. Device designed to carry out the method of one of 1-12. 14. A computer program, comprising instructions, configured, when executed on a computer processing unit, to perform a procedure according to one of 1-12. 15. A computer-readable medium comprising a computer program as defined in 14. Part 13E: Rover processing with ambiguity setting (A2599) [Summary (Fig. 49)] 1. A method for processing a set of GNSS signal data derived from signals of a set of satellites with carriers observed at a rover antenna, wherein the data include a carrier observation and a code observation of each carrier from each satellite, comprising: a. Obtain for each satellite time signal corrections comprising at least two of: (i) a code-matched satellite time signal, (ii) a phase-matched satellite time signal, and (iii) a satellite time signal bias representing a difference between a code-matched satellite time signal and a phase-matched satellite time signal. b. Running a first filter that uses at least the GNSS signal data and the satellite time signal corrections to estimate values ​​for parameters, comprising at least one carrier ambiguity for each satellite and a covariant matrix of carrier ambiguities, c. Determining each carrier ambiguity of a carrier ambiguity of an integer nature, comprising one of: an integer value, and a combination of integer candidate values, d. Inserting the carrier ambiguities of an integer nature as pseudo-observations into a second filter and applying the second filter to the GNSS signal data and the satellite time signal corrections to obtain estimated values ​​for parameters, including at least the receiver position. 2. The procedure according to 1, wherein the carrier ambiguities are of an integer nature and the intermediate satellites are individually differentiated ambiguities. 3. The procedure according to one of 1-2, furthermore comprehensive a. Obtaining a set of MW corrections, b. Running a third filter using the GNSS signal data and at least the MW corrections to obtain at least one set of WL ambiguities, and c. Using the set of WL ambiguities to preserve the carrier ambiguity of an integer nature. 4. The procedure according to 3, wherein the WL ambiguities include at least one of: floating values, integer values ​​and floating values ​​based on integer candidates. 5. The procedure according to 4, wherein the covariant matrix of ambiguities is scaled to reflect the change due to the use of the WL ambiguities. 6. The procedure according to one of 1-5, where the WL ambiguities are individually differentiated ambiguities for intermediate satellites. 7. The method according to one of 1-6, wherein the ambiguities of an integer nature include at least one of: L1-L2 ionosphere-free ambiguities, L2-L5 ionosphere-free ambiguities and carrier ambiguities of a linear combination of two or more GNSS frequencies. 8. The method according to one of 1-6, wherein the ionospheric delay information is used to feed to one or more of the filters, and wherein the ambiguity of an integer nature includes at least one of: a carrier ambiguity of the L1 frequency, carrier ambiguity of the L2 frequency, carrier ambiguity of the L5 frequency and carrier ambiguity of any GNSS frequency. 9. The method according to one of 1-8, wherein the second filter comprises one of: a new filter, a copy of the first filter and the first filter. 10. The method according to one of 1-9, wherein the code-matched satellite time signal is used to model all GNSS observations, and the floating ambiguity is formed to the level of the phase-matched time signal by applying the difference between the code-matched satellite time signal and the phase-matched satellite time signal. 11. The method according to one of 1-9, wherein the code-matched satellite time signal is used to model all GNSS code observations and the phase-matched satellite time signal is used to model all GNSS carrier observations. 12. The procedure according to one of 1-11, wherein determining the carrier ambiguity of an integer nature from a floating ambiguity comprises at least one of: rounding the floating ambiguity to the nearest integer, selecting best integer candidates from a set of integer candidates generated using integer least squares, and calculating floating values ​​using a set of integer candidates generated using integer least squares. 13. The method according to one of 1-12, wherein at least one of the first filter and the second filter further estimates at least one of: a receiver phase-matched time signal, receiver code-matched time signal, tropospheric delay, receiver time signal bias representing a difference between the code-matched receiver time signal and the phase-matched receiver time signal, and multipath states. 14. The method according to one of 1-13, wherein at least one of the first filter, the second filter and the third filter is configured to update the estimated values ​​for each of a plurality of epochs of GNSS signal data. 15. Device designed to carry out the method according to one of 1-14. 16. A computer program, comprising instructions, configured, when executed on a computer processing unit, to perform a procedure according to one of 1-14, 17. A computer-readable medium comprising a computer program as defined in 16.

[0393] Any of the aspects of the invention described above can be combined to form further aspects and embodiments with the aim of providing additional advantages, particularly with regard to measurement efficiency and / or system usability.

[0394] The methods, devices, and embodiments described above can be integrated into a rover, as well as a reference receiver or a network station, and / or the described processing methods can be executed in a processor that is separate from, and even remote from, the receiver(s) used to collect the observations (for example, observation data collected by one or more receivers can be retrieved from a memory for post-processing, or observations from multiple network reference stations can be transferred to a network processor for near real-time processing to generate a correction data stream and / or virtual reference station messages that can be transmitted to one or more rovers). Therefore, the invention also relates to a rover, a reference receiver, or a network station containing any of the above devices.

[0395] In some embodiments, the receiver of the device, as described above, is separate from the filter and the processing element. Post-processing and network processing of the observations can be performed. This means that the device's processing elements do not themselves require a receiver. The receiver can be separate from, and even owned / operated by, a different person or entity than the entity performing the processing. For post-processing, the observations can be retrieved from a previously collected and stored set of data and processed with previously collected and stored reference station data; the processing is performed, for example, on an office computer long after the data collection and is therefore not real-time.For network processing, multiple reference station receivers collect observations of signals from several satellites, and this data is fed to a network processor. This processor can, for example, generate a correction data stream or a "virtual reference station" correction, which is then fed to a rover so that the rover can perform differential processing. The data provided to the rover can be ambiguities determined by the network processor, which the rover can use to speed up its positioning solution, or it can be in the form of corrections that the rover applies to improve its positioning solution. The network is typically operated as a service for rover operators, while the network operator is typically a separate entity from the rover operator.

[0396] Each of the methods and embodiments described above can be implemented by means of a computer program. The computer program can be loaded into a device, a rover, a reference receiver, or a network station, as described above. Therefore, the invention also relates to a computer program which, when executed on a device, a rover, a reference receiver, or a network station, as described above, performs any one of the methods and embodiments described above.

[0397] The invention also relates to a computer-readable medium or a computer program product containing the aforementioned computer program. The computer-readable medium or computer program product can, for example, be a magnetic tape, an optical storage disk, a magnetic disk, a magneto-optical disk, a CD-ROM, a DVD, a CD, a flash memory device, or the like, wherein the computer program is stored permanently or temporarily. The invention also relates to a computer-readable medium (or a computer program product) containing computer-executable instructions for carrying out one of the methods of the invention.

[0398] The invention also relates to a firmware update designed to be installed on receivers already in the field, i.e., a computer program that has already been delivered to the field service or for practical use as a computer program product. This also applies to each of the methods and devices described above.

[0399] GNSS receivers may contain an antenna configured to receive signals at the frequencies transmitted by the satellites, processor units, one or more accurate time signals or clocks (such as crystal oscillators), one or more computer processing units (CPUs), one or more memory units (RAM, ROM, flash memory or similar), and a display to show position information to a user.

[0400] Where the terms "receiver," "filter," and "processing element" are used herein to refer to units of a device, no restriction is imposed as to how distributed the constituent parts of a unit may be. This means that the constituent parts of a unit may be distributed across different software or hardware components or devices to produce the desired function. Furthermore, the units may be combined to perform their functions by means of a single, combined unit. For example, the receiver, the filter, and the processing element may be combined to form a single unit for performing the combined functionalities of the units.

[0401] The units mentioned above can be implemented using hardware, software, a combination of hardware and software, pre-programmed ASICs (application-specific integrated circuits), etc. A unit can contain a computer processing unit (CPU), as well as a memory unit, input / output (I / O) units, network connectivity units, etc.

[0402] Although the present invention has been described on the basis of detailed examples, these detailed examples serve only to provide a better understanding to the person skilled in the art and are not intended to limit the scope of the invention. The scope of the invention is instead defined by the appended claims.

Claims

[1] A method for processing a set of GNSS signal data derived from signals of a set of satellites with carriers observed at a rover antenna, the data including a carrier observation and a code observation from each carrier of each satellite, comprising: Obtaining for each satellite time signal corrections comprising at least two of: (i) a code-adjusted satellite time signal, (ii) a phase-adjusted satellite time signal and (iii) a satellite time signal bias representing a difference between a code-adjusted satellite time signal and a phase-adjusted satellite time signal, Running a first filter that uses at least the GNSS signal data and the satellite time signal corrections to estimate values ​​for parameters comprising at least one carrier ambiguity for each satellite and a covariant matrix of the carrier ambiguities, wherein the code-adjusted satellite time signal is used to model all GNSS observations, and a sliding ambiguity is adjusted to the level of the phase-adjusted time signal by applying the difference between the code-adjusted satellite time signal and the phase-adjusted satellite time signal, Determining each carrier ambiguity of an integer nature comprising one of: an integer value, and a combination of integer candidates, inserting the carrier ambiguities of an integer nature as pseudo-observations into a second filter, and applying the second filter to the GNSS signal data and the satellite time signal corrections to obtain estimated values ​​for parameters comprising at least the position of the receiver. [2] The method of claim 1, wherein the carrier ambiguities are of integer nature and inter-satellite single-difference ambiguities. [3] The method of claim 1, further comprising Obtaining a set of MW corrections, Running a third filter using the GNSS signal data and at least the MW corrections to obtain at least one set of WL ambiguities, and Using the theorem of WL ambiguities to obtain the carrier ambiguity of integer nature. [4] The method of claim 3, wherein the WL ambiguities comprise at least one of: Floating values, integer values, and floating values ​​based on integer candidates. [5] The method of claim 4, wherein the covariant matrix of ambiguities is scaled to reflect the change due to the use of the WL ambiguities. [6] The method of claim 3, wherein the inter-satellite WL ambiguities are single-differenced ambiguities. [7] The method of claim 1, wherein the ambiguities of integer nature are at least one of: L1-L2 ionosphere-free ambiguities, L2-L5 ionosphere-free ambiguities and carrier ambiguities of a linear combination of two or more GNSS frequencies. [8] The method of claim 1, wherein the ionospheric delay information is used to supply one or more of the filters, and wherein the ambiguity of integer nature comprises at least one of: a carrier ambiguity of the L1 frequency, a carrier ambiguity of the L2 frequency, a carrier ambiguity of the L5 frequency, and a carrier ambiguity of any GNSS frequency. [9] The method of claim 1, wherein the second filter comprises one of: a new filter, a copy of the first filter, and the first filter. [10] The method of claim 1, wherein the code-adjusted satellite time signal is used to model all GNSS code observations and the phase-adjusted satellite time signal is used to model all GNSS carrier observations. [11] The method of claim 1, wherein determining the carrier ambiguity of an integer nature from a floating ambiguity comprises at least one of: rounding the floating ambiguity to the nearest integer, selecting best integer candidates from a set of integer candidates generated using integer least squares, and calculating floating values ​​using a set of integer candidates generated using integer least squares. [12] The method of claim 1, wherein at least one of the first filter and the second filter further estimates at least one of: a receiver phase-adjusted timing signal, receiver code-adjusted timing signal, tropospheric delay, receiver timing signal bias representing a difference between the code-adjusted receiver timing signal and the phase-adjusted receiver timing signal, and multipath conditions. [13] The method of claim 3, wherein at least one of the first filter, the second filter, and the third filter is configured to update the estimated values ​​for each of a plurality of epochs of GNSS signal data. [14] Apparatus for processing a set of GNSS signal data derived from signals of a first set of satellites having carriers observed at a rover antenna, the data including a carrier observation and a code observation from each carrier of each satellite, comprising: an element for obtaining, for each satellite, time signal corrections (4902), comprising at least two of: (i) a code-adjusted satellite time signal, (ii) a phase-adjusted satellite time signal, and (iii) a satellite time signal bias representing a difference between a code-adjusted satellite time signal and a phase-adjusted satellite time signal, a first filter (4916; 5016; 5118; 5218; 5318; 5418 5520) using at least the GNSS signal data and the satellite time signal corrections to estimate values ​​for parameters comprising at least one carrier ambiguity for each satellite and a covariant matrix of the carrier ambiguities, wherein the code-adjusted satellite time signal is used to model all GNSS observations, and the sliding ambiguity is adjusted to the level of the phase-adjusted time signal by applying the difference between the code-adjusted satellite time signal and the phase-adjusted satellite time signal, an element for determining each carrier ambiguity (4918) of a carrier ambiguity of an integer nature (4924), comprising one of: an integer value, and a combination of integer candidates, a second filter (4926; 5026; 5132; 5232; 5332; 5432; 5534; 5632; 5732) using the carrier ambiguities of an integer nature as pseudo-observations and using the GNSS signal data and the satellite time signal corrections to estimate values ​​for parameters comprising at least the position of the receiver. [15] The apparatus of claim 14, wherein the carrier ambiguities of an integer nature (4924) are inter-satellite single-difference ambiguities. [16] The device of claim 14, further comprising an element for obtaining a set of MW corrections, a third filter (5124; 5224; 5324; 5424; 5526) using the GNSS signal data and at least the MW corrections to obtain at least one set of WL ambiguities, and an element that uses the set of WL ambiguities to obtain the carrier ambiguity of integer nature. [17] The apparatus of claim 16, wherein the WL ambiguities comprise at least one of: Floating values, integer values, and floating values ​​based on integer candidates. [18] The apparatus of claim 17, wherein the covariant matrix of ambiguities is scaled to reflect the change due to the use of the WL ambiguities. [19] The apparatus of claim 16, wherein the inter-satellite WL ambiguities are single-difference ambiguities. [20] The apparatus of claim 14, wherein the ambiguities of integer nature comprise at least one of: L1-L2 ionosphere-free ambiguities, L2-L5 ionosphere-free ambiguities, and carrier ambiguities of a linear combination of two or more GNSS frequencies. [21] The apparatus of claim 14, wherein one or more of the filters is supplied with the ionospheric delay information and wherein the ambiguity of integer nature comprises at least one of: carrier ambiguity of the L1 frequency, carrier ambiguity of the L2 frequency, carrier ambiguity of the L5 frequency and carrier ambiguity of any GNSS frequency. [22] The apparatus of claim 14, wherein the second filter comprises one of: a new filter, a copy of the first filter, and the first filter. [23] The apparatus of claim 14, wherein the code-adjusted satellite time signal is used to model all GNSS code observations and the phase-adjusted satellite time signal is used to model all GNSS carrier observations. [24] The apparatus of claim 14, wherein the element for determining each carrier ambiguity of a carrier ambiguity of an integer nature (4924) is adapted to perform at least one of: rounding the floating ambiguity to the nearest integer, selecting best integer candidates from a set of integer candidates generated using integer least squares, and calculating floating values ​​using a set of integer candidates generated using integer least squares. [25] The apparatus of claim 14, wherein at least one of the first filter and the second filter is configured to further estimate at least one of: receiver phase-adjusted timing signal, receiver code-adjusted timing signal, tropospheric delay, receiver timing signal bias representing a difference between the code-adjusted receiver timing signal and the phase-adjusted receiver timing signal, and multipath conditions. [26] The apparatus of claim 16, wherein at least one of the first filter, the second filter, and the third filter is configured to update the estimated values ​​for each of a plurality of epochs of GNSS signal data. [27] A computer program comprising instructions configured, when executed on a computer processing unit, to carry out a method according to any one of claims 1-13. [28] A computer-readable medium comprising a computer program according to claim 27.

Citation Information

Patent Citations

  • Electrode Arrays and Methods of Fabrication Thereof

    US20090004471A1

  • Insulated conductor and method for manufacturing same

    US20090004472A1

  • Lyocell fibers

    US20090004473A1

  • Lyocell fibers with anti-microbial activity

    US20090004474A1

  • Conductive polymer composite structure

    US20090004476A1