A continuous carrier phase time transfer method for eliminating satellite arc segment day boundary jumps

Abstract

This invention belongs to the field of GNSS high-precision time and frequency transfer technology, and discloses a continuous carrier phase time transfer method to eliminate satellite arc segment date jumps. Based on satellite products with overlapping date epochs, this method employs an adjusted high-order Lagrange interpolation method. When the receiver observation epoch is at the date, the interpolation accurately extracts the date jump variables of adjacent product arc segments and uses them as correction terms for ambiguity parameters. This corrects error terms that vary depending on the satellite, eliminating the date jump in time comparison results caused by discontinuities in satellite product arc segments. By connecting adjacent satellite product arc segments using this method, the date jump system error contained in the satellite products is absorbed into the ambiguity parameters, avoiding contamination of receiver clock error estimation and other estimator states due to discontinuities in adjacent arc segments. This achieves continuous PPP carrier phase time transfer across time zones, improving time comparison accuracy and the continuity of the time link.

Description

Technical Field

[0001] This invention belongs to the field of GNSS high-precision time and frequency transfer technology, specifically relating to a continuous carrier phase time transfer method for eliminating satellite arc-segment solar boundary jumps. Background Technology

[0002] High-precision time and frequency transfer technologies for GNSS mainly include common-view, all-view, and carrier phase time transfer technologies based on precise point positioning (PPP). Common-view and all-view time transfer technologies are inherently limited by code pseudorange noise, restricting time transfer accuracy (nanosecond level) and short-term frequency stability. PPP time transfer technology improves time transfer accuracy to the sub-nanosecond level by introducing low-noise carrier phase observations. PPP technology relies on high-precision satellite orbit and clock error products provided by the International GNSS Service (IGS) to eliminate the influence of satellite-related errors. However, these precise satellite products are generated in daily batch estimates, and the solutions for adjacent arcs ("days" and "days") are independent. Due to differences in the accumulated daily average pseudorange noise and hardware delays within different arcs, satellite products for adjacent arcs exhibit discontinuities at the dateline (24:00:00 or 00:00:00). Furthermore, the arc configurations and model strategies adopted by different IGS analysis centers differ, resulting in inconsistent arc jump errors accumulated at the dateline for satellite products provided by different analysis centers. At the PPP time resolution user end, discontinuities in adjacent arc segments of satellite products introduce error terms at the date line that vary with the satellite. Time comparison users, based on different subsets of satellite observations, experience receiver clock errors and time comparison links that jump at the date line, known as date line discontinuities (DBDs). DBDs in PPP time transfer can cause time jumps of up to approximately 1 ns at the date line during time comparisons between two locations, significantly reducing time transfer accuracy and causing frequency drift, affecting both short-term and long-term frequency stability. This problem limits applications with stringent requirements for the continuity and long-term stability of the time comparison link, particularly advanced microwave and optical atomic clock time comparisons.

[0003] Because PPP relies on post-hoc precise satellite products, which are typically estimated in batches on a daily basis, discontinuities in satellite orbits and clock errors between adjacent arcs (days) are inevitable. When the receiver observation epoch crosses the dateline, the satellite product arc used needs to be switched, and the discontinuity error of the satellite product is introduced into the receiver clock error. Simultaneously, the application of higher-order interpolation causes this discontinuity error to propagate before and after the dateline epoch, resulting in significant satellite correlation errors at the dateline time, with an amplitude of up to 10 cm. This contaminates the PPP estimator state, causing DBDs (Dead-Divergence Disorders) in the receiver clock error and time comparison errors estimated at the dateline time. Currently, there are two main approaches to addressing the dateline discontinuity problem in the receiver clock error and time comparison estimated by PPP.

[0004] On the one hand, some scholars have attempted to directly stitch together the time link results, for example, by using ambiguity superposition and bidirectional filtering to achieve inter-day continuity. Although these methods alleviate the problem of discontinuity at the day boundary to some extent, they cannot effectively establish connections between adjacent arc segments because traditional satellite products do not overlap. Consequently, they cannot accurately obtain the error introduced by satellite products of adjacent arc segments at the day boundary. Therefore, at the day boundary, although these methods can effectively alleviate discontinuity, a time jump of about 0.15 ns still remains in the time comparison link.

[0005] On the other hand, other scholars have proposed an undifferentiated network solution method based on least-squares batch processing. This method selects a reference network and jointly utilizes observation data from multiple stations within the reference network over several consecutive days to uniformly estimate precise satellite clock errors. This generates precise satellite clock error products with continuous arc segments, replacing IGS products with discontinuities at the day boundary in adjacent arc segments, thus avoiding dependence on external IGS satellite products. However, this type of undifferentiated network solution method relies on the definition of an external clock error benchmark and requires the introduction of an additional complex satellite clock error estimation process, significantly increasing the implementation complexity and system burden for time comparison users.

[0006] In late 2024, some analysis centers at IGS attempted to extend the last epoch of the celestial (arc) satellite product from the original 23:59:30 or 23:45:00 to 24:00:00, making it overlap with the starting epoch 00:00:00 of the next arc, thus establishing a connection between adjacent satellite arcs. This improvement provided a new opportunity to study the discontinuity problem between adjacent arcs and demonstrated potential advantages in solving time alignment DBDs. However, existing PPP time transfer algorithms typically ignore the discontinuity between satellite product arcs. If existing algorithms are used directly, the overlapping 24:00:00 epoch will be discarded, failing to leverage its continuity advantage. Therefore, existing PPP time transfer methods lack a practical algorithm to effectively incorporate satellite products at the 24:00:00 epoch into time transfer, solve time link DBDs, and achieve continuity of the time alignment link at the solar boundary.

[0007] The statements in this section are merely background information related to the present invention and do not necessarily constitute prior art. Summary of the Invention

[0008] The purpose of this invention is to propose a continuous carrier phase time transfer method to eliminate satellite arc-segment date jumps. This method adjusts the ambiguity parameters at the date point to absorb satellite-related date discontinuity errors into the ambiguity parameters, thereby achieving a smooth transition of the cumulative delay of ambiguity parameters between adjacent arc segments. This ensures the continuity of receiver clock error during the date jump process, and thus achieves high-precision, continuous and reliable PPP time transfer.

[0009] To achieve the above objectives, the present invention adopts the following technical solution:

[0010] A continuous carrier phase time transfer method for eliminating satellite arc-segment diurnal jumps includes the following steps:

[0011] Step 1. Collect raw observation data from GNSS receivers connected to a high-precision external clock source, precise satellite orbit and satellite clock error products with overlapping epochs, and download auxiliary products of Earth rotation parameters;

[0012] Step 2. Based on the adjusted high-order Lagrange interpolation method, process the original observations epoch by epoch. Within the single satellite product arc to which the receiver observation epoch belongs, interpolate to obtain the precise satellite position vector and satellite clock error for that epoch.

[0013] When the receiver observes an epoch at the dateline, the epoch is considered to belong to two adjacent satellite product arcs at the same time. Interpolation is performed in the two arcs to obtain the precise satellite position vector and satellite clock error based on different arc products at that time.

[0014] Based on the two sets of precise satellite position vectors and satellite clock errors obtained by interpolation of adjacent product arcs at this epoch, the jump correction of adjacent arc satellite products at the date epoch is obtained by combining them.

[0015] Step 3. Construct a dual-frequency ionosphere-free combined observation equation based on the original pseudorange and carrier phase observations; correct the observations using the precise satellite position vector and satellite clock error obtained through interpolation; correct the propagation path error in the observation equation based on the model's empirical formula;

[0016] Step 4. Perform continuous parameter estimation based on Kalman filtering; when the receiver observation epoch is the date line, compensate for the ambiguity parameters based on the satellite product jump correction of adjacent arc segments, and correct the error terms caused by the discontinuity of product arc segments that differ with the satellite.

[0017] Two time users independently estimate the receiver clock difference and build a time comparison link based on the two clock difference results. During the clock difference differentiation process, the satellite product reference time included in the single-user clock difference result is eliminated to achieve direct comparison of the local time of the two users.

[0018] Furthermore, based on the aforementioned method for continuous carrier phase time transfer to eliminate satellite arc-segment date jumps, this invention also proposes a corresponding continuous carrier phase time transfer system for eliminating satellite arc-segment date jumps, the scheme of which is as follows:

[0019] A continuous carrier phase time transfer system for eliminating satellite arc-segment diurnal jumps includes the following modules:

[0020] The GNSS data preprocessing module is used to collect raw observation data from GNSS receivers connected to a high-precision external clock source, precise satellite orbit and satellite clock error products with overlapping epochs, and download auxiliary products of Earth rotation parameters.

[0021] A module for precise satellite position and clock error interpolation and extraction of satellite product jump corrections for adjacent arc segments at the date horizon; Based on an adjusted high-order Lagrange interpolation method, the module processes the original observations epoch by epoch, and within the single satellite product arc segment to which the receiver observation epoch belongs, it interpolates to obtain the precise satellite position vector and satellite clock error for that epoch;

[0022] When the receiver observes an epoch at the dateline, the epoch is considered to belong to two adjacent satellite product arcs at the same time. Interpolation is performed in the two arcs to obtain the precise satellite position vector and satellite clock error based on different arc products at that time.

[0023] Based on the two sets of precise satellite position vectors and satellite clock errors obtained by interpolation of adjacent product arcs at this epoch, the jump correction of adjacent arc satellite products at the date epoch is obtained by combining them.

[0024] The error processing and observation equation establishment module is used to construct a dual-frequency ionosphere-free combined observation equation based on the original pseudorange and carrier phase observations; correct the observations using the precise satellite position vector and satellite clock error obtained by interpolation; and correct the propagation path error in the observation equations based on model empirical formulas.

[0025] And a receiver parameter estimation and time link comparison module to eliminate the discontinuity of the solar boundary, used for continuous parameter estimation based on Kalman filtering; when the receiver observation epoch is the solar boundary time, the ambiguity parameters are compensated based on the jump correction of satellite products in adjacent arc segments, and the error terms that differ with satellites due to the discontinuity of product arc segments are corrected.

[0026] Two time users independently estimate the receiver clock difference and build a time comparison link based on the two clock difference results. During the clock difference differentiation process, the satellite product reference time included in the single-user clock difference result is eliminated to achieve direct comparison of the local time of the two users.

[0027] Furthermore, based on the aforementioned method for continuous carrier phase time transfer to eliminate satellite arc-level diurnal jumps, this invention also proposes a computer device comprising a memory and one or more processors.

[0028] Executable code is stored in memory. When the processor executes the executable code, it implements the steps of the above-described method for continuous carrier phase time transfer to eliminate satellite arc-level diurnal jumps.

[0029] Furthermore, based on the above-mentioned method for continuous carrier phase time transfer to eliminate satellite arc-segment diurnal jumps, this invention also proposes a computer-readable storage medium storing a program that, when executed by a processor, implements the steps of the above-mentioned method for continuous carrier phase time transfer to eliminate satellite arc-segment diurnal jumps.

[0030] The present invention has the following advantages:

[0031] As described above, this invention discloses a method for eliminating the impact of satellite arc segment boundary jumps and realizing continuous carrier phase time transfer based on PPP. This method is based on precise satellite orbits and clock bias products with overlapping boundary epochs. It employs an adjusted high-order Lagrange interpolation method, interpolating the satellite orbits and clock bias products of adjacent arc segments at the receiver observation time as the boundary epoch, thereby accurately extracting the jump correction amount of satellite products in adjacent arc segments. Unlike the continuous estimation of carrier phase ambiguity in traditional PPP time transfer, this invention corrects the ambiguity parameters based on this jump correction amount at the receiver observation time as the boundary epoch. This is used to compensate for the error term introduced by the discontinuity of satellite product arc segments at the boundary epoch, which varies with the satellite, thereby eliminating the boundary discontinuity of the estimated receiver clock bias. This method is simply referred to as Ambiguity Correction on Overlapping-boundary Epoch (ACOE). By correcting the ambiguity parameters at the day boundary time to absorb the jumps in satellite products of adjacent arc segments, the system can effectively and accurately connect adjacent satellite arc segments of the satellite products used. This absorbs the error terms caused by the discontinuity of satellite product arc segments into the ambiguity, effectively avoiding contamination of receiver clock bias estimation and other estimator states. This improves the robustness of the estimator, enables continuous PPP time transfer across days, and significantly enhances the accuracy of time comparison and the continuity of the time link. Attached Figure Description

[0032] Figure 1 This is a flowchart of the continuous carrier phase time transfer method for eliminating satellite arc-segment solar boundary jumps in an embodiment of the present invention;

[0033] Figure 2This is a diagram of discrete overlapping dateline precision satellite products and adjusted Lagrange interpolation curves in an embodiment of the present invention;

[0034] Figure 3 A schematic diagram of the receiver clock bias of the TAI laboratory AOS station AO_5 for traditional PPP and ACOE PPP solutions;

[0035] Figure 4 A diagram showing the AOS-PL time link results for epoch-differential traditional PPP and the ACOE PPP of this invention;

[0036] in Figure 4 In the diagram, (a) represents the epoch differential time solution of the AOS station AO_5 receiver using the traditional PPP and ACOE PPP methods. Figure 4 (b) in the table represents the time-link AOS-PL epoch differential time solution for traditional PPP and ACOE PPP methods;

[0037] Figure 5 The diagram shows the time link results of long baseline PTB-NIST for epoch-difference traditional PPP, the ACOE PPP of this invention, and the International Weights Bureau PPP. Figure 5 (a) in the text compares traditional PPP with BIPM, and (b) compares ACOE with BIPM. Detailed Implementation

[0038] The present invention will be further described in detail below with reference to the accompanying drawings and specific embodiments:

[0039] Example 1

[0040] This embodiment 1 provides a method for eliminating the impact of satellite arc segment date jumps and achieving continuous carrier phase PPP time transfer. This method corrects the ambiguity parameters at the date line, absorbing the error term introduced by the discontinuity of satellite product arc segments, which varies with the satellite, into the ambiguity parameters. This achieves a smooth transition of the cumulative delay of ambiguity parameters between adjacent arc segments, thereby ensuring the continuity of the receiver clock error during the time-crossing process and achieving high-precision, continuous, and reliable PPP time transfer.

[0041] The method of this invention includes steps such as GNSS data preprocessing, precise satellite position and clock error interpolation and extraction of jump corrections for satellite products in adjacent arc segments, error processing and observation equation establishment, receiver parameter estimation and time link comparison.

[0042] like Figure 1 As shown, the continuous carrier phase time transfer method for eliminating satellite arc-segment diurnal jumps includes the following steps:

[0043] Step 1. GNSS data preprocessing.

[0044] Collect raw observation data from GNSS receivers connected to a high-precision external clock source, as well as precise satellite orbit and satellite clock bias products with overlapping epochs. Download auxiliary products of Earth rotation parameters and perform quality control on the raw observation data.

[0045] In this embodiment, satellite orbit and clock bias products with overlapping epochs provided by the German Research Centre for Geosciences (GFZ) are downloaded, along with Earth rotation parameter products provided by the IGS organization. RINEX files corresponding to pseudorange and carrier phase observations from a GNSS receiver connected to a high-precision external clock source are obtained, and epoch-by-epoch processing is performed on the GNSS observation data.

[0046] The quality control check is performed on the dual-frequency observations at the current epoch, the gross errors of the pseudorange observations are eliminated, the cycle slip is detected on the carrier phase observations, and abnormal or incomplete satellite observation data are eliminated.

[0047] Step 2. Precise satellite position and clock error interpolation and extraction of satellite product jump corrections for adjacent arc segments.

[0048] The raw GNSS receiver observation data is processed epoch-by-epoch using an adjusted high-order Lagrange interpolation method. Within a single arc of the satellite product to which the receiver observation epoch belongs, the precise satellite position vector and satellite clock error at that epoch are obtained through interpolation.

[0049] During the interpolation process, the adjusted higher-order Lagrange interpolation in this invention adopts an interpolation window of fixed length, and all product interpolation nodes within the window should be in the same satellite product arc segment as the receiver observation epoch, i.e., on the same day, to avoid the day boundary jump variable of adjacent arc segments being diffused to epochs before and after the day boundary time through higher-order interpolation.

[0050] Meanwhile, satellite products located at overlapping epoch times should be retained as effective interpolation nodes. This is to avoid wasting information from precise satellite products and to accurately extract jump corrections for satellite products in adjacent arc segments based on these overlapping epoch times.

[0051] Furthermore, when the receiver observation epoch is the dateline, the precise satellite position and satellite clock error results obtained from the latest arc interpolation are used for observation correction. Specifically, the higher-order Lagrange interpolation that adjusts the GNSS satellite orbit and clock error products at discrete times is shown in Equation (1);

[0052] (1)

[0053] in The receiver observation time is the first Heaven and the second The precise satellite position or satellite clock error obtained by interpolation, in meters; For the first Lagrange basis functions for interpolation nodes; This is the order of the Lagrange interpolation.

[0054] Indicates the first [item / product name] provided by precision satellite products Heaven and the second Time Satellite The satellite position or satellite clock bias.

[0055] and Indicates the first element participating in the interpolation. The and the first The day corresponding to a discrete satellite product at a given time; and Indicates the first element participating in the interpolation. The and the first The second within a day corresponding to a discrete moment in a satellite product. and These represent the indices of the interpolation nodes.

[0056] Indicates satellite product interpolation nodes and The selection should be limited to the same day as the receiver's observation epoch. ; This is a rough estimate of the satellite signal propagation time. and This refers to the day and second within the day for the receiver to observe the epoch time.

[0057] Satellite orbital product interpolation uses the 10th order, i.e. Satellite clock bias products adopt first-order... , .

[0058] During the precise satellite product estimation process, satellite products in adjacent arc segments accumulate different hardware delays and daily average noise, resulting in discontinuities in satellite products in adjacent arc segments.

[0059] Furthermore, when the receiver observes an epoch at the International Date Line, this epoch can be considered to belong to two adjacent satellite product arcs simultaneously. By interpolating the precise satellite orbit and satellite clock bias products with overlapping International Date Line epochs within the two product arcs, precise satellite position vectors and satellite clock biases based on interpolation of products from different arcs can be obtained at that moment. Based on the two sets of precise satellite position vectors and satellite clock biases obtained from interpolation of adjacent product arcs at that epoch, the difference between the two sets of precise satellite clock biases (or satellite position vectors) can be obtained to determine the jump variables of the adjacent arc segment satellite clock bias (or satellite orbit) products at that moment. .

[0060] like Figure 2 This displays satellite products at discrete time points and interpolation curves based on the adjusted higher-order Lagrange method. The blue and orange lines represent the results of the adjusted higher-order Lagrange interpolation. Tianhe Di The satellite position or clock error curves are displayed, and the interpolation results are magnified at the date (00:00:00) to show the jump variables of satellite clock error (or satellite orbit) products in adjacent arc segments at that time. Specifically, as shown in formula (2);

[0061] (2)

[0062] in For the receiver observation epoch, the epoch is the 1st. Satellites with Lagrange interpolation at the boundary time Jump variables in satellite orbits (components) and clock bias products for adjacent arc segments; considering that the precise satellite position vector is composed of various coordinate components, , , These correspond to the components of the satellite orbit product jump variable in the X, Y, and Z directions for adjacent arc segments at that time, respectively, while the satellite clock difference product jump variable for adjacent arc segments at that time is expressed as... ; and The first Tianhe Di The set of Lagrange interpolation nodes for the celestial satellite arc segment; The first for precision satellite products Heaven and the second Time Satellite Satellite position or satellite clock bias; Indicates satellite For the first At 0:00 on the day and night, the first second Lagrange basis functions for interpolation nodes; For satellite No. Heaven and the second Time of the first Lagrange basis functions for interpolation nodes.

[0063] In the ionospheric-free combined observation equation, the satellite-receiver geometric distance is expressed by the distance formula between two points. The satellite orbit product jump variable of adjacent arc segments at the receiver observation epoch of the date line is projected onto the satellite-receiver line-of-sight direction. The processing is as follows: In order to explicitly describe the influence of the discontinuous jump of the satellite position vector between adjacent arc segments on the geometric distance, and to obtain the projection correction value of the satellite orbit product jump variable between adjacent arc segments in the receiver-satellite direction, the satellite-receiver geometric distance expression is Taylor expanded on the satellite position vector, as shown in equation (3):

[0064] (3)

[0065] In the formula The satellite position vector is the one observed by the receiver at the International Date Line. The satellite orbital jump variable, defined by the receiver observation time at the epoch of the solar calendar, is composed of orbital jump components in the X, Y, and Z directions at that time. ; and These are respectively derived from satellite position vectors and Satellites determined at time to receiver geometric distance; The receiver-to-satellite unit vector; It is the identity matrix; Expand Taylor series terms of third order and above;

[0066] When the receiver observation time is at the solar epoch, the projection of the satellite orbit product jump variable of adjacent arc segments onto the receiver-satellite direction can be expressed as: Because this jump variable is much smaller than the geometric distance between the satellite and the receiver. Its quadratic term And higher-order terms are ignored; therefore, the jump correction for satellite orbit products in adjacent arc segments is ;

[0067] Furthermore, the correction for the jump in satellite products in adjacent arc segments at the date horizon consists of correction terms for the jump variables in satellite orbit products and satellite clock difference products, expressed as formula (4):

[0068] (4)

[0069] in This invention provides jump corrections for satellite products in adjacent arc segments when the receiver observation epoch is the date. Utilizing precise satellite orbit and clock bias products with overlapping date epochs, the invention employs the ACOE method to accurately extract jump corrections for satellite products in adjacent arc segments at the date. Furthermore, when interpolating precise satellite position vectors and clock biases using an adjusted higher-order Lagrange interpolation method, the interpolation is confined to the same product arc segment, effectively avoiding the problem of discontinuity in satellite product arc segments spreading through higher-order interpolation in traditional PPP methods.

[0070] This invention obtains the satellite product jump correction amount for adjacent arc segments using formula (4) when the receiver observation epoch is the date horizon. Unlike the traditional continuous estimation of PPP ambiguity parameters, this invention adds an adjusted high-order Lagrange interpolation method and ambiguity parameter correction if and only if the receiver observation epoch is the date horizon. ), based on the jump correction of satellite products in adjacent arc segments Correct the floating-point ambiguity parameter.

[0071] The ambiguity parameter absorbs the influence of satellite products accumulated in adjacent arc segments, ensuring the stability of the carrier phase equation residuals, avoiding the estimator being contaminated by error terms that vary with different satellites at the date, and solving the problem of date discontinuity in receiver clock bias.

[0072] Furthermore, unlike traditional methods that splice time link results to alleviate date discontinuities, the ACOE method addresses satellite-related arc discontinuities, achieving cross-day continuity of estimated receiver clock biases and thus guaranteeing cross-day continuous PPP time alignment from the source. Moreover, unlike multi-day continuous network solutions based on reference networks that eliminate date discontinuities through continuous satellite clock estimation, the ACOE method of this invention does not require defining a reference network clock bias benchmark or performing additional satellite clock bias estimation, significantly reducing the implementation complexity for time users.

[0073] Step 3. Error processing and establishment of observation equations.

[0074] A dual-frequency, ionospherically-free combined observation equation is constructed based on the original pseudorange and carrier phase observations from the GNSS receiver to eliminate the effects of first-order ionospheric delay. The observations are corrected using the precise satellite position vector and satellite clock bias obtained from interpolation in step 2. Furthermore, the satellite-related, receiver-related, and atmospheric-related errors in the observations are corrected based on empirical formulas from the model, such as relativistic effects, Earth's rotation, tropospheric delay, antenna phase entanglement, antenna phase center, and tidal effects.

[0075] For the portion of the propagation path error that is difficult to correct using the model, continuous parameter estimation is performed using Kalman filtering of the GNSS observation equations. Based on the obtained satellite correlation jump corrections for adjacent arc segments, when the receiver observation epoch is the dateline, the ambiguity parameters to be estimated are corrected, enabling a smooth and reliable transition of ambiguity parameters that have accumulated the influence of satellite arc segments.

[0076] Therefore, the satellite product jump correction for adjacent arc segments of the solar epoch is introduced into the observation equations of the parameters to be estimated, and the observation equations are shown in equations (5), (6) and (7):

[0077] (5)

[0078] (6)

[0079] (7)

[0080] In the formula and These represent pseudorange and carrier phase observations, respectively, based on the elimination of the first-order ionospheric delay effect by combining dual-frequency signals. The floating-point ambiguity to be estimated. and This represents the raw pseudorange observations of satellite signals at different frequencies output by the receiver; and This represents the original carrier phase observations of satellite signals at different frequencies output by the receiver; and This represents the combination coefficient of the dual-frequency non-ionospheric combination. Indicates satellite to receiver geometric distance; The receiver clock bias to be estimated; Precision satellite clock biases provided by the International GNSS Service (IGS) organization; The receiver observation epoch is expressed in seconds within a day. and This is for dual-frequency, ionospherically-free combined pseudorange and carrier phase observation noise; This is for tropospheric delay.

[0081] In formula (7) This means that the receiver observes an epoch that is the International Date Line. The ambiguity parameter has been redefined, namely, the correction amount for satellite product jumps in adjacent arc segments at that moment. Correct the floating-point ambiguity parameter.

[0082] It should be noted that pseudorange observations are two orders of magnitude lower than carrier phase observations (typical values ​​are 0.3 m and 3 mm, respectively), and the jump variables of satellite products in adjacent arcs (typical value is 0.2 m) will be directly absorbed by the pseudorange residuals in the pseudorange observation equation.

[0083] The symbols in formula (7) , indicating if and only if the time of day (i.e., the time of day) This formula redefines the ambiguity parameter, namely, the jump correction based on adjacent arc segments of satellite products. Correct the floating-point ambiguity parameter.

[0084] Similarly, to iteratively solve for the receiver's precise position vector, the satellite-receiver geometric distance expression in observation equations (5) and (6) is derived based on formula (3). Perform a Taylor expansion on the known receiver position vector.

[0085] The linearization of the GNSS observation equation and the variance matrix of the parameters to be estimated are written as Equation (8) and Equation (9), respectively.

[0086] (8)

[0087] In the formula For vector residuals, The coefficient matrix of the parameters to be estimated. The vector of parameters to be estimated A constant term vector;

[0088] ; ; ;

[0089] ;

[0090] In the formula The number of observation satellites; and For the pseudorange and carrier phase residuals of satellite number 1 among the user-observable satellites; and Among the user-observable satellites, the one numbered is... Satellite pseudorange and carrier phase residuals. and These are the pseudorange and carrier phase observations without ionospheric combination after deducting the error term. and For satellite number 1 among the user-observable satellites, the ionospheric combined pseudorange and carrier phase observations are obtained by correcting signal propagation path errors based on model empirical formulas. and Among the user-observable satellites, numbered The satellite provides ionospherically unbound pseudorange and carrier phase observations based on model empirical formulas to correct signal propagation path errors.

[0091] Tropospheric delay in formulas (5) and (6) The delay is decomposed into an oblique tropospheric dry delay and an oblique tropospheric wet delay. The oblique tropospheric dry delay is corrected using the GPT2w and Saastamoinen models, while the oblique tropospheric wet delay is represented as the zenith tropospheric wet delay. and tropospheric wet delay mapping function .

[0092] The tropospheric wet delay mapping function uses GPT2w and GMF, and the zenith wet delay estimation is a random walk process.

[0093] The increment of the receiver coordinates to be estimated. and For user-observable satellites numbered 1 and The tropospheric wet delay mapping function for satellites. and For user-observable satellites numbered 1 and The floating-point ambiguity of the satellite to be estimated. This represents the position vector of satellite number 1 among the user-observable satellites; This indicates the number of user-observable satellites. Satellite position vector; This is the receiver position vector known during this iteration. This represents the receiver-satellite number 1 geometric distance calculated based on the satellite position vector and the receiver position vector; Represents the receiver number calculated based on the satellite position vector and the receiver position vector. The geometric distance of the satellite.

[0094] definition Let be the variance matrix of the parameters to be estimated, and let the initial variance of the location be set to . The position parameter is estimated to be constant, and the initial variance of the receiver clock bias is... And estimated as white noise, the initial variance of the carrier phase ambiguity parameter is set to The continuous locked observation arc is estimated to be constant, and the initial variance of the zenith tropospheric delay is... It is estimated to be a random walk process, as shown in formula (9).

[0095] (9)

[0096] In this embodiment, the following values ​​can be selected for example: , , , .

[0097] Step 4. Receiver parameter estimation and time link comparison.

[0098] Continuous parameter estimation is performed based on Kalman filtering, without the need for segmented processing by day. When the receiver observation epoch is the date, the estimated ambiguity parameter is corrected based on the obtained jump correction of satellite products in adjacent arc segments (Formula (7)) to achieve a smooth transition of the cumulative delay of ambiguity parameters in adjacent arc segments, thereby ensuring the continuity of receiver clock error in the cross-day process.

[0099] Two time users independently estimate the receiver clock bias. The receiver clock bias estimated by a single time user represents the deviation of the local time from the satellite product reference time. A time comparison link is constructed based on the clock bias results of the two time users. During the clock bias differential process, the satellite product reference time included in the single user clock bias result is eliminated, thus enabling direct comparison of the local times of the two users.

[0100] This invention employs epoch-by-epoch Kalman filtering to continuously estimate the parameters to be estimated, including the receiver coordinate increment vector, receiver clock bias, zenith tropospheric wet delay, and carrier phase ambiguity parameter vector. Based on an adjusted high-order Lagrange interpolation method, satellite orbit and clock bias interpolation is limited to the same product arc segment, preventing the dilatational jump variables of adjacent arc segments from being diffused to epochs before and after the dilatational time through high-order interpolation. By correcting the ambiguity parameters at the dilatational time, the jumps in satellite products of adjacent arc segments are absorbed, thereby achieving effective and accurate connection of adjacent satellite arc segments for the satellite products used. This ensures that the error term caused by the discontinuity of satellite product arc segments, which varies with the satellite, is absorbed into the ambiguity parameters, guaranteeing continuous dilatational estimation of the receiver clock bias during the dilatational process.

[0101] A time comparison link is constructed based on the continuous clock difference results across two time users. During the clock difference differentiation process, the satellite product reference time included in the clock difference results of a single user is eliminated, thereby directly obtaining the time difference between the local times of the two users.

[0102] To verify the performance of the method of this invention, a GNSS receiver station connected to the local UTC(k) in the International Atomic Time (TAI) laboratory was selected. The receiver clock bias was calculated and the time between the two locations was compared based on the traditional PPP method and the ACOE method adopted in this invention. The comparison and analysis were conducted to see whether the ACOE method can achieve continuous receiver clock bias estimation, and thus achieve continuous time comparison.

[0103] like Figure 3This paper demonstrates the clock bias of a receiver at the Astrogeodynamical Observatory of the Space Research Centre (AOS) TAI laboratory, calculated using both the conventional PPP and the PPP time transfer method with Overlapping Ephemeris Ambiguity Correction (ACOE) (ACOE PPP) employed in this invention. During the experiment, the clock of the AOS TAI laboratory GNSS receiver AO_5 remained continuously running without any clock jumps or restarts. Figure 3 The results demonstrate this situation in the receiver clock bias estimated by the ACOE PPP scheme. Furthermore, the ACOE method resolves the jump in receiver clock bias estimated by the traditional PPP method at the day's date, achieving a continuous-time solution.

[0104] Epoch difference is often used to detect the diurnal jump of receiver clock errors. Figure 4 The results are shown after performing epoch difference on the time comparison link results obtained by traditional PPP and the ACOE PPP method used in this invention.

[0105] Figure 4 In the diagram, (a) represents the epoch differential time solution of the AOS station AO_5 receiver using the traditional PPP and ACOE PPP methods. Figure 4 In the example (b), the time-link AOS-PL epoch difference time solution is shown for both traditional PPP and ACOE PPP methods.

[0106] Figure 3 The diurnal transition of the AO_5 receiver clock bias is clearly displayed. Figure 4 In (a), the traditional PPP method receiver clock error solution absorbs the transitions of satellite products in adjacent arc segments, while the ACOE PPP method extracts the transition corrections of satellite products in adjacent arc segments and eliminates this effect. The AO_5 receiver clock error of the epoch difference is maintained at ±0.05 ns, realizing a continuous-time solution.

[0107] Furthermore, a short-baseline time link is formed by connecting AOS and the PL laboratory (Consortium of laboratories in Poland) in Warszawa, Poland. Figure 4 (b) in the diagram shows the link epoch difference results.

[0108] Because AOS-PL is a short baseline time link, and the AOS and PL stations contain a consistent subset of observed satellites, the traditional PPP method mitigates the diurnal discontinuity of the AO_5 receiver clock bias during time alignment (receiver clock bias differential). However, a diurnal jump exceeding 0.05 ns still exists (marked by the green box). In contrast, the ACOE PPP method achieves reliable cross-day continuous time alignment by addressing the diurnal jump in the receiver clock bias.

[0109] Figure 5 The results are shown after performing epoch differencing on the long baseline link time comparison results obtained by traditional PPP, the ACOE PPP method adopted in this invention, and the International Bureau of Weights and Measures (BIPM).

[0110] Figure 5 (a) shows a comparison of results between traditional PPP and BIPM, and (b) shows a comparison of results between the present invention's ACOE and BIPM. PTB-NIST (7532.2 km) represents the long baseline time alignment experimental link connecting the German Federal Institute of Physics and Technology (PTB) and the National Institute of Standards and Technology (NIST) in the United States, with a time baseline length of approximately 7532.2 km.

[0111] In the long baseline time comparison experiment, the PTB and NIST stations had completely different subsets of observed satellites at the same time, therefore Figure 5 In (a) of the diagram, the traditional PPP method failed to mitigate the impact of the date jump through time comparison, resulting in a jump at the date. The ACOE method, however, addresses the date jump caused by the PTB and NIST receiver clock differences, achieving continuous time comparison across days. Furthermore, compared to the BIPM PPP scheme, the PTB-NIST time comparison results obtained by the ACOE PPP method remain consistent after epoch differencing, and neither scheme's time comparison results show a jump at the date. It should be noted that outside the date, both schemes exhibit the same jump in the epoch differencing results for this time link. Combining this with the NIST laboratory NISG receiver clock difference results provided by the BIPM FTP platform (ftp: / / ftp2.bipm.org / pub / tai / timelinks / lkc / ), a consistent jump was found, suggesting that the intraday jump was primarily caused by the local clock.

[0112] The above results demonstrate that the Overlapping Ephemeris Ambiguity Correction (ACOE) proposed in this invention, unlike traditional PPP methods which weaken time link time link time jumps through time comparison, addresses the estimated receiver clock bias time jump, thus ensuring reliable PPP time comparison across consecutive days from the source. The PPP time transfer method based on Overlapping Ephemeris Ambiguity Correction (ACOE) proposed in this invention can accurately extract and compensate for the error term introduced by the discontinuity of satellite products in adjacent arc segments, resulting in differences in satellite characteristics, and achieves continuous and stable estimation of the local receiver clock bias across consecutive days. Existing PPP time transfer methods generally ignore this error term and its generation mechanism, weakening the impact of time link discontinuities (such as ambiguity overlap, bidirectional filtering, and data overlap) by splicing daily time link results; however, a time link time jump of approximately 0.15 ns still remains in the time comparison link.

[0113] Furthermore, unlike multi-day continuous network solutions based on reference networks that eliminate discontinuities at the day boundary through continuous satellite clock estimation, this invention eliminates the need to define a reference network clock bias benchmark and perform additional satellite clock bias estimation, significantly reducing the implementation complexity for time users. This invention effectively avoids the problem of discontinuities in adjacent arc segments propagating within higher-order interpolation windows and failing to be absorbed by the estimator parameters, preventing such errors from contaminating the estimator state and causing aberrant posterior residuals and discarding observational information. By processing satellite-related errors introduced by adjacent arc segments at the day boundary time, the continuity and stability of the posterior residuals are ensured, thereby fully preserving effective observational information and achieving continuous and reliable estimation of the estimator parameters.

[0114] In addition, compared with daily batch processing, the multi-day continuous processing ACOE method proposed in this invention keeps the receiver clock error of the date time estimation continuous, providing reliable technical support for high-precision time comparison and continuous time measurement.

[0115] Example 2

[0116] This embodiment 2 describes a continuous carrier phase time transfer system for eliminating satellite arc-segment solar boundary jumps. This system is based on the same inventive concept as the continuous carrier phase time transfer method for eliminating satellite arc-segment solar boundary jumps in embodiment 1.

[0117] The continuous carrier phase time transfer system for eliminating satellite arc-segment solar term jumps in this embodiment includes the following modules:

[0118] A continuous carrier phase time transfer system for eliminating satellite arc-segment diurnal jumps includes the following modules:

[0119] The GNSS data preprocessing module is used to collect raw observation data from GNSS receivers connected to a high-precision external clock source, precise satellite orbit and satellite clock error products with overlapping epochs, and download auxiliary products of Earth rotation parameters.

[0120] A module for precise satellite position and clock error interpolation and extraction of satellite product jump corrections for adjacent arc segments at the date horizon; Based on an adjusted high-order Lagrange interpolation method, the module processes the original observations epoch by epoch, and within the single satellite product arc segment to which the receiver observation epoch belongs, it interpolates to obtain the precise satellite position vector and satellite clock error for that epoch;

[0121] When the receiver observes an epoch at the dateline, the epoch is considered to belong to two adjacent satellite product arcs at the same time. Interpolation is performed in the two arcs to obtain the precise satellite position vector and satellite clock error based on different arc products at that time.

[0122] Based on the two sets of precise satellite position vectors and satellite clock errors obtained by interpolation of adjacent product arcs at this epoch, the jump correction of adjacent arc satellite products at the date epoch is obtained by combining them.

[0123] At this point, the precise satellite position and satellite clock error results obtained from the latest arc interpolation are used to correct the observation values.

[0124] The error processing and observation equation establishment module is used to process GNSS receiver observation data epoch by epoch and construct observation equations for user-end parameters to be estimated; construct dual-frequency ionosphere-free combined observation equations based on the original pseudorange and carrier phase observations; correct the observations using the precise satellite position vector and satellite clock error obtained by interpolation; and correct the propagation path error in the observation equations based on model empirical formulas.

[0125] And a receiver parameter estimation and time link comparison module to eliminate the discontinuity of the solar boundary, used for continuous parameter estimation based on Kalman filtering; when the receiver observation epoch is the solar boundary time, the ambiguity parameters are compensated based on the jump correction of satellite products in adjacent arc segments, and the error terms that differ with satellites due to the discontinuity of product arc segments are corrected.

[0126] Two time users independently estimate the receiver clock difference and build a time comparison link based on the two clock difference results. During the clock difference differentiation process, the satellite product reference time included in the single-user clock difference result is eliminated to achieve direct comparison of the local time of the two users.

[0127] It should be noted that any content not mentioned in the above-described functional modules of the system described in Embodiment 2 can be referred to the step description of the corresponding method in Embodiment 1 above, and will not be repeated in detail here.

[0128] Example 3

[0129] This embodiment 3 describes a computer device including a memory and one or more processors. Executable code is stored in the memory. When the processor executes the executable code, it implements the steps of the continuous carrier phase time transfer method for eliminating satellite arc date jumps in embodiment 1 above.

[0130] In this embodiment, the computer device can be any device or apparatus with data processing capabilities, and will not be described in detail here.

[0131] Example 4

[0132] This embodiment 4 describes a computer-readable storage medium storing a program that, when executed by a processor, is used to implement the steps of the continuous carrier phase time transfer method for eliminating satellite arc-level diurnal jumps in embodiment 1 above.

[0133] The computer-readable storage medium can be an internal storage unit of any device or apparatus with data processing capabilities, such as a hard disk or memory, or an external storage device of any device with data processing capabilities, such as a plug-in hard disk, smart media card (SMC), SD card, flash card, etc.

[0134] Of course, the above description is only a preferred embodiment of the present invention. The present invention is not limited to the above-described embodiments. It should be noted that any equivalent substitutions or obvious modifications made by those skilled in the art under the guidance of this specification fall within the scope of this specification and should be protected by the present invention.

Claims

1. A continuous carrier phase time transfer method for eliminating satellite arc-segment date jumps, characterized in that, Includes the following steps: Step 1. Collect raw observation data from GNSS receivers connected to a high-precision external clock source, precise satellite orbit and satellite clock error products with overlapping epochs, and download auxiliary products of Earth rotation parameters; Step 2. Based on the adjusted high-order Lagrange interpolation method, process the original observations epoch by epoch. Within the single satellite product arc to which the receiver observation epoch belongs, interpolate to obtain the precise satellite position vector and satellite clock error for that epoch. When the receiver observes an epoch at the dateline, the epoch is considered to belong to two adjacent satellite product arcs at the same time. Interpolation is performed in the two arcs to obtain the precise satellite position vector and satellite clock error based on different arc products at that time. Based on the two sets of precise satellite position vectors and satellite clock errors obtained by interpolation of adjacent product arcs at this epoch, the jump correction of adjacent arc satellite products at the date epoch is obtained by combining them. Step 3. Construct a dual-frequency ionospheric-free combined observation equation based on the original pseudorange and carrier phase observations; correct the observations using the precise satellite position vector and satellite clock error obtained by interpolation; Step 4. Perform continuous parameter estimation based on Kalman filtering. When the receiver observation epoch is the date line, compensate for the ambiguity parameters based on the jump correction of satellite products in adjacent arc segments to correct the error terms caused by the discontinuity of product arc segments that differ from satellite presentation. Two time users independently estimate the receiver clock difference and build a time comparison link based on the two clock difference results. During the clock difference differentiation process, the satellite product reference time included in the single-user clock difference result is eliminated to achieve direct comparison of the local time of the two users.

2. The continuous carrier phase time transfer method for eliminating satellite arc-segment date jumps according to claim 1, characterized in that, In step 1, the raw observation data is subjected to quality control, including gross error detection of pseudorange observations, cycle slip detection of carrier phase observations, and removal of incomplete or abnormal satellite observation data.

3. The continuous carrier phase time transfer method for eliminating satellite arc-segment date jumps according to claim 1, characterized in that, In step 2, the GNSS satellite orbit and clock error products at discrete times are interpolated based on the adjusted high-order Lagrange interpolation method to obtain the precise satellite position vector and satellite clock error for that observation epoch. During the interpolation process, the adjusted higher-order Lagrange interpolation uses a fixed-length interpolation window, and all product interpolation nodes within the window should be in the same satellite product arc segment as the receiver observation epoch, i.e., within the same day; Meanwhile, satellite products located at overlapping epoch times should be retained as valid interpolation nodes; When the receiver observation epoch is at the dateline, the epoch is considered to belong to two adjacent satellite product arcs at the same time. In this case, the precise satellite position and satellite clock error results obtained by the latest arc interpolation are used for observation correction.

4. The continuous carrier phase time transfer method for eliminating satellite arc-segment date jumps according to claim 3, characterized in that, The adjusted higher-order Lagrange interpolation formula is shown in Equation (1); （1） in The receiver observation time is the first Heaven and the second The precise satellite position or satellite clock error obtained by interpolation, in meters; For the first Lagrange basis functions for interpolation nodes; Let Lagrange be the interpolation order; Indicates the first [item / product name] provided by precision satellite products Heaven and the second Time Satellite Satellite position or satellite clock bias; and Indicates the first element participating in the interpolation. The and the first The day corresponding to a discrete satellite product at a given time; and Indicates the first element participating in the interpolation. The and the first The second within a day corresponding to a discrete moment in a satellite product. and These represent the interpolation node indices; Indicates satellite product interpolation nodes and The selection should be limited to the same day as the receiver's observation epoch. ; This is a rough estimate of the satellite signal propagation time. and For the receiver to observe the epoch time in days and seconds within days; Satellite orbital product interpolation uses the 10th order, i.e. Satellite clock bias products adopt first-order... , .

5. The continuous carrier phase time transfer method for eliminating satellite arc-segment date jumps according to claim 4, characterized in that, In step 2, when the receiver observation epoch is at the solar boundary, this epoch is considered to belong to two adjacent satellite product arcs simultaneously. Based on satellite products with overlapping solar boundary epochs, an adjusted high-order Lagrange interpolation formula is used. When the receiver observation epoch is at the solar boundary, interpolation is performed in the adjacent product arcs before and after, respectively, to obtain two sets of precise satellite position vectors and satellite clock errors. The two sets of precise satellite clock errors or satellite position vectors are differentiated to obtain the satellite clock error or satellite orbit product jump variable of the adjacent arc at that moment. (2) in For the receiver observation epoch, the epoch is the 1st. Satellites with Lagrange interpolation at the boundary time Jump variables in satellite orbit components and clock bias products for adjacent arc segments; considering that the precise satellite position vector is composed of various coordinate components. , , These correspond to the components of the satellite orbit product jump variable in the X, Y, and Z directions for adjacent arc segments at that time, respectively, while the satellite clock difference product jump variable for adjacent arc segments at that time is expressed as... ; and The first Tianhe Di The set of Lagrange interpolation nodes for the celestial satellite arc segment; The first for precision satellite products Heaven and the second Time Satellite Satellite position or satellite clock bias; Indicates satellite For the first At 0:00 on the day and night, the first second Lagrange basis functions for interpolation nodes; For satellite No. Heaven and the second Time of the first Lagrange basis functions for interpolation nodes.

6. The continuous carrier phase time transfer method for eliminating satellite arc-segment date jumps according to claim 5, characterized in that, In step 2, the satellite orbit product jump variables of adjacent arc segments when the receiver observation epoch is the date boundary are projected onto the satellite-receiver line-of-sight direction. The processing procedure is as follows: In order to explicitly describe the impact of the discontinuous jump of the satellite position vector between adjacent arc segments on the geometric distance, and to obtain the projection correction value of the satellite orbit product jump variables between adjacent arc segments in the receiver-satellite direction, the satellite-receiver geometric distance expression is Taylor expanded on the satellite position vector, as shown in equation (3): (3) In the formula The satellite position vector is the one observed by the receiver at the International Date Line. The satellite orbital jump variable, defined by the receiver observation time at the epoch of the solar calendar, is composed of orbital jump components in the X, Y, and Z directions at that time. ; and These are respectively derived from satellite position vectors and Satellites determined at time to receiver geometric distance; The receiver-to-satellite unit vector; It is the identity matrix; Expand Taylor series terms of third order and above; When the receiver observation time is at the solar epoch, the projection of the satellite orbit product jump variable of adjacent arc segments onto the receiver-satellite direction can be expressed as: Because this jump variable is much smaller than the geometric distance between the satellite and the receiver. Its quadratic term And higher-order terms are ignored; therefore, the jump correction for satellite orbit products in adjacent arc segments is ; The correction for satellite product jumps in adjacent arc segments at the date horizon consists of correction terms for jumps in satellite orbit products and jumps in satellite clock products, expressed as formula (4): (4) in This is the jump correction for satellite products in adjacent arc segments when the receiver's observation epoch is the date.

7. The continuous carrier phase time transfer method for eliminating satellite arc-segment date jumps according to claim 6, characterized in that, In step 3, a satellite product jump correction for adjacent arc segments of the epoch of the day is introduced into the observation equation of the parameter to be estimated. The observation equations are shown in equations (5), (6) and (7): (5) (6) (7) In the formula and These represent pseudorange and carrier phase observations, respectively, based on the elimination of the first-order ionospheric delay effect by combining dual-frequency signals. The floating-point ambiguity to be estimated; and This represents the raw pseudorange observations of different frequency satellite signals output by the receiver; and This represents the original carrier phase observations of satellite signals at different frequencies output by the receiver; and This represents the combination coefficient of the dual-frequency non-ionospheric component; Indicates satellite to receiver geometric distance; The receiver clock bias to be estimated; Precision satellite clock bias provided by the International GNSS Service (IGS) organization; For receiver observation epochs, expressed in seconds within a day; and This is for dual-frequency, ionospherically-free combined pseudorange and carrier phase observation noise; For tropospheric delay; In formula (7) This means that the receiver observes an epoch that is the International Date Line. The ambiguity parameter has been redefined, namely, the correction amount for satellite product jumps in adjacent arc segments at that moment. Correct the floating-point ambiguity parameter.

8. The continuous carrier phase time transfer method for eliminating satellite arc-segment date jumps according to claim 7, characterized in that, In step 3, to iteratively solve for the receiver's precise position vector, the satellite-receiver geometric distance expression in observation equations (5) and (6) is derived based on formula (3). Perform a Taylor expansion on the known receiver position vector; The linearization of the GNSS observation equation and the variance matrix of the parameters to be estimated are written as Equation (8) and Equation (9), respectively. (8) In the formula For vector residuals, The coefficient matrix of the parameters to be estimated. The vector of parameters to be estimated A constant term vector; ； ； ； ； In the formula The number of observation satellites; and For the pseudorange and carrier phase residuals of satellite number 1 among the user-observable satellites; and Among the user-observable satellites, numbered as Satellite pseudorange and carrier phase residuals; and For satellite number 1 among the user-observable satellites, the ionospheric combined pseudorange and carrier phase observations are obtained by correcting the signal propagation path error based on model empirical formulas. and Among the user-observable satellites, numbered Satellites, with ionospheric pseudorange and carrier phase observations corrected for signal propagation path errors based on model empirical formulas; For zenith tropospheric wet delay; and For user-observable satellites numbered 1 and The floating-point ambiguity of the satellite to be estimated; The receiver coordinate increment to be estimated; and For user-observable satellites numbered 1 and The tropospheric wet delay mapping function for satellites; This represents the position vector of satellite number 1 among the user-observable satellites; This indicates the number of user-observable satellites. Satellite position vector; The receiver position vector is known during this iteration; This represents the receiver-satellite number 1 geometric distance calculated based on the satellite position vector and the receiver position vector; Represents the receiver number calculated based on the satellite position vector and the receiver position vector. The geometric distance of the satellite; definition Let be the variance matrix of the parameters to be estimated, and let the initial variance of the location be set to . The position parameter is estimated to be constant, and the initial variance of the receiver clock bias is... And estimated as white noise, the initial variance of the carrier phase ambiguity parameter is set to The continuous locked observation arc is estimated to be constant, and the initial variance of the zenith tropospheric delay is... It is estimated to be a random walk process, as shown in formula (9); (9) 。 9. The continuous carrier phase time transfer method for eliminating satellite arc-segment date jumps according to claim 1, characterized in that, In step 4, Kalman filtering is used to estimate the parameters to be estimated without the need for daily segmentation. The parameters to be estimated include receiver coordinate increment vector, receiver clock error, zenith tropospheric wet delay, and carrier phase ambiguity parameter vector.

10. A continuous carrier phase-time transfer system for eliminating satellite arc-segment date jumps, characterized in that, Includes the following modules: The GNSS data preprocessing module is used to collect raw observation data from GNSS receivers connected to a high-precision external clock source, precise satellite orbit and satellite clock error products with overlapping epochs, and download auxiliary products of Earth rotation parameters. The module for precise satellite position and clock error interpolation and extraction of satellite product jump corrections in adjacent arc segments at the date boundary is used to process the original observations epoch by epoch based on the adjusted high-order Lagrange interpolation method. Within the single satellite product arc segment to which the receiver observation epoch belongs, the module interpolates to obtain the precise satellite position vector and satellite clock error of that epoch. When the receiver observes an epoch at the dateline, the epoch is considered to belong to two adjacent satellite product arcs at the same time. Interpolation is performed in the two arcs to obtain the precise satellite position vector and satellite clock error based on different arc products at that time. Based on the two sets of precise satellite position vectors and satellite clock errors obtained by interpolation of adjacent product arcs at this epoch, the jump correction of adjacent arc satellite products at the date epoch is obtained by combining them. The error processing and observation equation establishment module is used to construct a dual-frequency ionosphere-free combined observation equation based on the original pseudorange and carrier phase observations; correct the observations using the precise satellite position vector and satellite clock error obtained by interpolation; and correct the propagation path error in the observation equations based on model empirical formulas. And a receiver parameter estimation and time link comparison module to eliminate the discontinuity of the solar boundary, used for continuous parameter estimation based on Kalman filtering; when the receiver observation epoch is the solar boundary time, the ambiguity parameters are compensated based on the jump correction of satellite products in adjacent arc segments, and the error terms that differ with satellites due to the discontinuity of product arc segments are corrected. Two time users independently estimate the receiver clock difference and build a time comparison link based on the two clock difference results. During the clock difference differentiation process, the satellite product reference time included in the single-user clock difference result is eliminated to achieve direct comparison of the local time of the two users.

Images