Carrier phase cycle slip real-time detection and repair method for airborne GNSS receiver

Abstract

The application discloses a carrier phase cycle slip real-time detection and repair method of an airborne GNSS receiver, and relates to the technical field of satellite navigation and positioning. GNSS signal data is acquired; phase variation of adjacent ephemeris is calculated, and a cycle slip detection model is constructed; a multi-dimensional filtering method is used to smooth the carrier phase data, so as to reduce measurement noise; in combination with satellite geometric information, a non-difference and double-difference method is used to correct errors; a prediction method based on Kalman filtering or an autoregressive model is used to repair the detected cycle slip data; finally, baseline solution results are recalculated, quality control is performed, and it is ensured that the repaired GNSS data meet high-precision positioning requirements. The method can effectively improve the positioning precision of the airborne GNSS receiver in a high dynamic environment, enhance the integrity and reliability of data, and has important engineering application value.

Description

TECHNICAL FIELD

[0001] The present application relates to the field of satellite navigation and positioning technology, and in particular to a carrier phase cycle slip real-time detection and repair method for an airborne GNSS receiver. BACKGROUND

[0002] GNSS (Global Navigation Satellite System) has been widely used in the fields of aviation navigation, precise measurement and scientific research. However, due to factors such as signal blockage, ionospheric delay change, and receiver hardware jitter, GNSS carrier phase observations are prone to sudden changes, which is called carrier phase cycle slip. Cycle slip will reduce the positioning accuracy of the global navigation satellite system, and even cause the failure of the navigation satellite system. Therefore, it is necessary to detect and repair the carrier phase cycle slip in real time to reduce the impact on the positioning accuracy.

[0003] The traditional cycle slip detection method is mainly based on a simple threshold method of phase mutation between single epochs, which is difficult to deal with error accumulation in complex environments. In addition, the traditional carrier phase cycle slip repair method often relies on high-precision auxiliary data, such as reference stations or multi-frequency observations, which is not applicable to the application scenario of airborne GNSS receivers that cannot rely on fixed base stations. SUMMARY

[0004] The present application proposes a carrier phase cycle slip real-time detection and repair method for an airborne GNSS receiver to solve the technical problem that the traditional cycle slip detection method is difficult to deal with error accumulation in complex environments and is not applicable to airborne GNSS receivers. Efficient and robust cycle slip detection and repair are achieved.

[0005] In a first aspect, the embodiments of the present application provide a carrier phase cycle slip real-time detection and repair method for an airborne GNSS receiver, comprising: acquiring GNSS signal data in real time, the GNSS signal data comprising carrier phase observations; determining an initial phase change between adjacent epochs based on the carrier phase observations;

[0006] The cycle slip detection model is constructed, and the initial phase change quantity is input as an input parameter into the cycle slip detection model to perform preliminary cycle slip detection; the carrier phase observation value is smoothed, denoised and enhanced, and the error of the carrier phase observation value is corrected to obtain a corrected carrier phase observation value; the corrected carrier phase observation value is used to determine a corrected phase change quantity; the corrected phase change quantity is input as an input parameter into the cycle slip detection model to obtain initial cycle slip data; the initial cycle slip data includes a satellite in which cycle slip occurs and an epoch in which cycle slip occurs; the initial cycle slip data is repaired to obtain cycle slip data, and a repaired carrier phase observation value is obtained based on the cycle slip data; and the baseline solution result is re-determined based on the repaired carrier phase observation value, and quality control is performed on the repaired carrier phase observation value.

[0007] In combination with the first aspect, in a possible implementation manner, the phase change quantity includes: ; in the formula, represents a phase change quantity between adjacent epochs, represents a carrier phase observation value of the first k epoch, represents a carrier phase observation value of the first k -1 epoch.

[0008] In combination with the first aspect, in a second possible implementation manner, the preliminary cycle slip detection includes: setting a cycle slip threshold value, and calculating a mean value and a standard deviation of the initial phase change quantity; based on the initial phase change quantity, the mean value, the standard deviation and the cycle slip threshold value, the epoch range in which cycle slip occurs is preliminarily determined; a sudden change point is obtained based on wavelet transform; and the epoch range in which cycle slip occurs is verified or supplemented through the sudden change point.

[0009] In combination with the second possible implementation manner of the first aspect, in a third possible implementation manner, the verification or supplement of the epoch range in which cycle slip occurs based on the wavelet transform includes: calculating a local sudden change feature of the initial phase change quantity based on the wavelet transform to obtain a sudden change point; and verifying or supplementing the epoch range in which cycle slip occurs based on the sudden change point.

[0010] In combination with the first aspect, in a fourth possible implementation manner, the smoothing, denoising and enhancement processing of the carrier phase observation value includes: performing data smoothing, denoising and enhancement processing on the carrier phase observation value based on a multi-dimensional filtering method, including: predicting a current state of the carrier phase observation value by a conventional Kalman filtering method, and calculating a Kalman gain of the carrier phase observation value; performing smoothing processing on the carrier phase observation value based on the current state and the Kalman gain; denoising the smoothed carrier phase observation value by a median filtering method; and performing signal enhancement on the denoised carrier phase observation value by a Wiener filter.

[0011] In a fifth possible implementation manner of the first aspect, the error of the carrier phase observation is corrected, and corrected carrier phase observation is obtained, including: calculating the carrier phase observation of the first satellite at the first epoch based on a carrier phase observation equation; calculating the pseudo-range observation of the first satellite at the first epoch based on a pseudo-range observation equation; calculating the error range of the carrier phase based on an error propagation formula; and correcting the error range based on the pseudo-range observation to obtain the corrected carrier phase observation. i The satellite in the first k epoch; calculating the pseudo-range observation of the first satellite at the first i epoch based on a pseudo-range observation equation; calculating the error range of the carrier phase based on an error propagation formula; and correcting the error range based on the pseudo-range observation to obtain the corrected carrier phase observation. k

[0012] In a sixth possible implementation manner of the first aspect, the initial cycle slip data is repaired to obtain cycle slip data, including: repairing the initial cycle slip data based on a regression prediction method, including: filling in the missing values in the initial cycle slip data by using a linear interpolation method to obtain filled cycle slip data; and establishing a dynamic repair model based on a recursive least squares algorithm, and inputting the filled cycle slip data as input parameters into the dynamic repair model to optimize the repair result to obtain the cycle slip data.

[0013] In a seventh possible implementation manner of the sixth possible implementation manner of the first aspect, after the cycle slip data is obtained, the method further includes: inputting the filled cycle slip data and the cycle slip data as training samples into a long short-term memory network learning model for training; and predicting and repairing the filled cycle slip data in other epochs based on the trained long short-term memory network learning model to obtain predicted and repaired cycle slip data.

[0014] In an eighth possible implementation manner of the first aspect, the baseline solution result is recalculated based on the repaired carrier phase observation, including: determining the carrier phase of the repaired carrier phase observation based on a carrier phase observation equation, and determining the pseudo-range observation of the repaired carrier phase observation based on a pseudo-range observation equation; constructing a double-difference observation equation based on the carrier phase and the pseudo-range observation, and eliminating common error terms; the common error terms include receiver clock errors and satellite clock errors; estimating ambiguity parameters in the double-difference observation equation based on a least squares method; determining the baseline solution result according to the ambiguity parameters and the double-difference observation equation; the baseline solution result includes a baseline vector and a covariance matrix of the baseline vector; wherein the baseline vector represents the relative position relationship between two airborne GNSS receivers, and the covariance matrix represents the accuracy of the baseline solution result.

[0015] ​With reference to the eighth possible implementation manner of the first aspect, in a ninth possible implementation manner, the quality control on the repaired carrier phase observation value comprises: obtaining a residual error of the pseudo-range observation value based on the receiver clock error and the pseudo-range observation value in the repaired carrier phase observation value; determining a standard deviation based on the residual error of the pseudo-range observation value; calculating a reliability factor through the residual error of the pseudo-range observation value and the standard deviation; obtaining a final positioning error based on a position solution accuracy; and feeding back to the repaired carrier phase observation value by using the residual error, the reliability factor and the final positioning error, so as to control the quality of the repaired carrier phase observation value.

[0016] The one or more technical solutions provided in the embodiments of the present application have at least the following technical effects or advantages:

[0017] The present application provides a carrier phase cycle slip real-time detection and repair method for an airborne GNSS receiver, mainly aiming at the carrier phase cycle slip phenomenon caused by the influence of various factors (such as high dynamic motion, ionospheric delay, shielding, etc.) on GNSS signals in an airborne environment, and proposes a complete cycle slip detection, data smoothing, error correction, cycle slip repair and quality control method to improve the continuity and accuracy of GNSS measurement data. BRIEF DESCRIPTION OF DRAWINGS

[0018] In order to more clearly illustrate the technical solutions of the embodiments of the present application, the following will briefly introduce the drawings needed to be used in the embodiments of the present application or the prior art description. Obviously, the drawings in the following description are some embodiments of the present application, and those skilled in the art can obtain other drawings according to these drawings without creative labor.

[0019] Figure 1 A flow chart of a carrier phase cycle slip real-time detection and repair method for an airborne GNSS receiver provided by the embodiments of the present application;

[0020] Figure 2 A carrier phase change curve provided by the embodiments of the present application;

[0021] Figure 3 A cycle slip detection schematic diagram provided by the embodiments of the present application;

[0022] Figure 4 A comparison curve diagram before and after cycle slip repair provided by the embodiments of the present application. DETAILED DESCRIPTION

[0023] With reference to the drawings of the embodiments of the present application, the technical solutions in the embodiments of the present application will be described clearly and completely. Obviously, the described embodiments are part of the embodiments of the present application, rather than all the embodiments. Based on the embodiments in the present application, all other embodiments obtained by those of ordinary skill in the art without creative work fall within the protection scope of the present application.

[0024] Figure 1 A flowchart of a carrier phase cycle slip real-time detection and repair method of an airborne GNSS receiver provided by the embodiments of the present application is shown in FIG. 1, and the method provided by the embodiments of the present application includes steps S1 to S7. Figure 1

[0025] Step S1: Real-time acquisition of GNSS signal data, wherein the GNSS signal data includes carrier phase observation value, pseudorange, satellite azimuth and other observation data.

[0026] Step S2: Calculation of initial phase change between adjacent epochs based on the carrier phase observation value.

[0027] Step S3: Construction of a cycle slip detection model, and input of the initial phase change as an input parameter into the cycle slip detection model for preliminary cycle slip detection.

[0028] Step S4: Smoothing, denoising and enhancement processing of the carrier phase observation value, and correction of errors of the carrier phase observation value; obtaining of corrected carrier phase observation value, and calculation of corrected phase change based on the corrected carrier phase observation value.

[0029] Step S5: Input of the corrected phase change as an input parameter into the cycle slip detection model to obtain initial cycle slip data; the initial cycle slip data includes a satellite with cycle slip and an epoch with cycle slip.

[0030] Step S6: Repair of the initial cycle slip data to obtain cycle slip data, and obtaining of repaired carrier phase observation value based on the cycle slip data.

[0031] Step S7: Recalculation of baseline solution results based on the repaired carrier phase observation value, and quality control of the repaired carrier phase observation value.

[0032] The present application provides a carrier phase cycle slip real-time detection and repair method of an airborne GNSS receiver, mainly aiming at the carrier phase cycle slip phenomenon caused by the influence of various factors (such as high dynamic motion, ionospheric delay, shielding, etc.) on GNSS signals in an airborne environment, and proposes a complete cycle slip detection, data smoothing, error correction, cycle slip repair and quality control method to improve the continuity and accuracy of GNSS measurement data. ​

[0033] Although the present application provides method operation steps as described in the embodiments or flowcharts, more or less operation steps can be included based on routine or non-creative labor. The order of steps listed in the embodiments is only one of the many ways of execution order, and does not represent the only execution order. In actual device or client product execution, the method order shown in the embodiments or the drawings can be executed in sequence or in parallel (for example, in a parallel processor or multi-threaded processing environment).

[0034] It should be noted that in step S1, GNSS signal data transmitted by satellites needs to be received based on an airborne GNSS receiver, and frequency conversion, amplification, filtering and digitization are performed thereon. The GNSS signal data includes carrier phase observations, pseudoranges, satellite azimuths and other observation data. Among them, the airborne GNSS receiver includes a single-frequency receiver, a multi-frequency receiver and a high-dynamic receiver.

[0035] The signal receiving process includes signal acquisition and signal tracking.

[0036] Among them, the signal acquisition is used to identify visible satellites and preliminarily lock the signals. The principle is that the spread spectrum signal transmitted by the satellite includes a pseudo-random noise code and a carrier; the receiver performs correlation operation on the received signal by generating a local pseudo-code, and captures the signal after successful matching.

[0037] And the signal tracking is used to continuously lock the satellite signals to extract the pseudorange observations and the carrier phase observations. The change of the carrier phase observations is mainly tracked by a phase-locked loop. The phase change of the pseudorange observations is tracked by a delay-locked loop.

[0038] Exemplarily, the initial phase change amount mentioned in step S2 is calculated as follows:

[0039] .

[0040] In the formula, represents the initial phase change amount between adjacent epochs, represents the initial carrier phase observation of the same satellite at the k epoch, represents the initial carrier phase observation of the same satellite at the k -1 epoch. The phase change amount can be used for high-precision positioning. It should be noted that the phase change amount between adjacent epochs represents the change of the carrier phase observation of the same satellite between two consecutive time points, and the purpose is to detect cycle slips. In the embodiments of the present application, the initial phase change amount is only used for preliminary cycle slip detection (judging the approximate epoch range of the cycle slip occurrence).

[0041] Before preliminary cycle slip detection, a cycle slip detection model needs to be constructed. In the cycle slip detection model, the initial phase change is used as the input parameter to perform preliminary cycle slip detection.

[0042] Specifically, the preliminary cycle slip detection process includes:

[0043] Set a cycle slip threshold and calculate the mean and standard deviation of the initial phase change.

[0044] The calculation method for the mean of the initial phase change includes:

[0045] .

[0046] In the formula, μ This represents the mean. N Indicates the number of epochs within the time window. Indicates the first k -1 epoch and the first k The initial phase change between epochs, that is, the initial phase change between two adjacent epochs.

[0047] The standard deviation of the initial phase change is calculated in the following ways:

[0048] .

[0049] In the formula, It represents the square of the standard deviation. It represents the standard deviation.

[0050] Based on the initial phase change, mean, standard deviation, and cycle slip threshold, the epoch range in which the cycle slip occurred is preliminarily determined, specifically including:

[0051] .

[0052] In the formula, This represents a pre-set cycle slip threshold. When the condition of this formula is met, it indicates that the cycle slip threshold is... k A cycle jump occurred.

[0053] In the embodiments of this application, the method of using wavelet transform to verify or supplement the epoch range in which cycle slips occur specifically includes:

[0054] The abrupt change characteristics of the initial phase change are calculated based on wavelet transform to obtain the abrupt change point. Specifically, this includes:

[0055] .

[0056] In the formula, Describes the wavelet transform function. Describe the wavelet basis functions.a Indicates scale. b Indicates the translation parameter. t This represents the abrupt change point of the wavelet coefficients, i.e., the epoch in which the cycle slip occurs. Indicates in t The carrier phase observation value at time t.

[0057] Based on the mutation point, verify or supplement the epoch range where cycle slips occur.

[0058] In practical applications, carrier phase measurements face two key challenges: measurement noise and cycle slips. Measurement noise is high-frequency random noise caused by multipath effects, ionospheric disturbances, and receiver hardware errors. Cycle slips, on the other hand, are abrupt changes in carrier phase integer cycles, such as those caused by signal obstruction, which disrupt phase continuity and require timely detection and repair. Therefore, subsequent calculations necessitate the use of multidimensional filtering methods to smooth, denoise, and enhance carrier phase observations. Multidimensional filtering utilizes information from multiple dimensions, such as time, frequency, space, and satellite ensemble distribution, to construct a filtering model and jointly process the original phase changes.

[0059] Specifically, the carrier phase observations are smoothed, denoised, and enhanced, including:

[0060] Step 401: Predict the current state of the carrier phase observations based on conventional Kalman filtering, including:

[0061] .

[0062] In the formula, Indicates the first k The current state of the carrier phase observation at an epoch, i.e., the state prediction vector. A Represents the state transition matrix. Indicates the first k -1 epoch state prediction vector B Represents the control input matrix. This represents the control input vector.

[0063] Step 402: Calculate the Kalman gain of the carrier phase observations to smooth the carrier phase observations between adjacent epochs based on the current state and the Kalman gain, removing high-frequency noise and preserving the true trend of phase change. This includes:

[0064] .

[0065] In the formula, Indicates the first k Kalman gain matrix of epochs, H Represents the observation matrix. T Indicates transpose.R covariance of the observation noise, covariance of the prediction state of the first k epoch, k covariance matrix of the prediction state of the first , covariance of the process noise, covariance of the prediction state of the first k epoch.

[0066] Step 403: denoising the smoothed carrier phase observations by median filtering. Median filtering effectively suppresses noise interference by sorting the values within a window of time series data and replacing the value of the center point with the median value.

[0067] Step 404: enhancing the denoised carrier phase observations by Wiener filtering. Wiener filtering minimizes the mean square error by optimizing the filter coefficients based on the statistical characteristics of the signal and noise, thereby improving the signal quality. It includes:

[0068] .

[0069] wherein, denotes the carrier phase observations after Wiener filtering, denotes the power spectral density of the carrier phase observations, denotes the power spectral density of the noise, denotes the observed signal, , denotes the frequency domain of the carrier phase observations, denotes the frequency domain of the noise.

[0070] In the embodiments of the present application, the error of the carrier phase observations is corrected as described in step S4 to obtain corrected carrier phase observations. The error is corrected by non-difference and double-difference methods, wherein the non-difference method directly uses the original carrier phase observations of a single receiver for a single satellite without differential processing. The double-difference method differentiates the carrier phase observations of the same group of satellites observed by two receivers (reference stations and mobile stations) twice to eliminate the single difference between reference stations, satellite clock error, inter-satellite single difference, and receiver clock error. Specifically, it includes:

[0071] Step 405: calculating the carrier phase observations of the second i satellite at the first k epoch based on the carrier phase observation equation, which includes:

[0072] .

[0073] wherein, denotes the carrier phase observations of the second iThe satellite in the k Carrier phase observations at each epoch. Indicates the first i The satellite in the k Geometric distance of epochs, Indicates the carrier wavelength. Indicates the first i The satellite in the k The ambiguity parameter of the epoch. Indicates the first i The satellite in the k Atmospheric delay of an epoch, Indicates the first i The satellite in the k Epochal observation error.

[0074] Step 406: Calculate the first based on the pseudorange observation equation i The satellite in the k The pseudorange observations at each epoch. The pseudorange observation equations include:

[0075] .

[0076] In the formula, Indicates the first i The satellite in the k pseudorange observations at epochs, c Represents the speed of light. Indicates that the receiver is at the first i The receiver clock bias when a satellite receives signals.

[0077] Step 407: Calculate the error range of the carrier phase using the error propagation formula. Specifically, this includes:

[0078] .

[0079] In the formula, The standard deviation of the carrier phase error. D Represents the design matrix. Represents the error covariance matrix. T This indicates transpose.

[0080] Step 408: Establish a joint observation model and use pseudorange observations to assist in correcting the error range, obtaining the corrected carrier phase observations. This includes:

[0081] .

[0082] In the formula, This indicates the auxiliary correction result, where, This represents the amount of correction change in the carrier phase. This represents the amount of correction change for pseudorange observations. a design matrix representing carrier phases, a design matrix representing pseudo-range observations, x a state vector, an observation noise matrix, wherein, v a weight parameter, a weight parameter of a carrier phase observation of a satellite at an epoch. i a weight parameter of a carrier phase observation of a satellite at an epoch. k

[0083] Based on the auxiliary correction result, the error range of the carrier phase observation is corrected to obtain a corrected carrier phase observation.

[0084] After obtaining the corrected carrier phase observation, the corrected phase change amount is calculated by the method as recorded in step S2.

[0085] After obtaining the corrected phase change amount, the corrected phase change amount is taken as an input parameter of a cycle slip detection model to obtain initial cycle slip data, wherein the initial cycle slip data includes a satellite i and an epoch k in which cycle slip occurs.

[0086] The initial cycle slip data is repaired to obtain cycle slip data, and a repaired carrier phase observation is obtained based on the cycle slip data.

[0087] Specifically, the process of repairing the initial cycle slip data includes a prediction method based on Kalman filtering or an autoregressive model to repair the initial cycle slip data, and the specific process is as follows:

[0088] A linear interpolation method is used to fill in missing values in the initial cycle slip data to obtain filled cycle slip data to restore the continuity of the initial cycle slip data. Including:

[0089] .

[0090] In the formula, denotes a carrier phase observation at an epoch, k denotes a carrier phase observation at an epoch, denotes a carrier phase observation at an epoch, k denotes a carrier phase observation at an epoch. k A dynamic repair model is established by a recursive least squares algorithm, and the filled cycle slip data is taken as an input parameter and input into the dynamic repair model to optimize the repair result for the initial cycle slip data to obtain cycle slip data.

[0091]

[0092] . ​​​

[0093] wherein, denotes the state estimation vector at the epoch, k denotes the state estimation vector at the epoch, denotes the state estimation vector at the epoch, k denotes the state estimation vector at the epoch, denotes the Kalman gain matrix, denotes the observation vector, denotes the observation matrix at the epoch, k denotes the observation matrix at the epoch.

[0094] The dynamic repair model optimizes the repair result of the initial cycle slip data step by step through iterative processing, and improves the data accuracy of the cycle slip data.

[0095] After obtaining the repaired cycle slip data, a long short-term memory network learning model (LSTM model, Long Short-Term Memory) is selected, and the filled cycle slip data and the cycle slip data are input into the long short-term memory network learning model as training samples for training.

[0096] Based on the trained long short-term memory network learning model, the filled cycle slip data in other epochs is predicted and repaired to obtain the predicted and repaired cycle slip data. Specifically, it includes:

[0097] .

[0098] wherein, y denotes the predicted and repaired cycle slip data, f is a prediction function of the machine learning model, x is an input feature vector (i.e., a state vector).

[0099] Meanwhile, the baseline solution result is recalculated based on the repaired carrier phase observation value in step S7 described in the embodiments of the present application, specifically including:

[0100] Step 701: Calculate the carrier phase of the repaired carrier phase observation value based on the carrier phase observation equation, and calculate the pseudo-range observation value of the repaired carrier phase observation value based on the pseudo-range observation equation.

[0101] Step 702: Based on the carrier phase and the pseudo-range observation value, a double-difference observation equation is constructed, the carrier phase and the pseudo-range observation value are combined in double difference, and common error terms are eliminated; the common error terms include receiver clock error and satellite clock error.

[0102] Step 703: Estimate the ambiguity parameter in the double-difference observation equation based on the least square method or other estimation method.

[0103] Step 704: calculating a baseline solution according to the ambiguity parameters and the double-difference observation equation; the baseline solution includes a baseline vector and a covariance matrix of the baseline vector, wherein the baseline vector represents a relative position relationship between two airborne GNSS receivers, and the covariance matrix represents an accuracy of the baseline solution.

[0104] In the embodiments of the present application, the quality control of the repaired carrier phase observation value described in step S7 includes:

[0105] Step 705: obtaining a residual of the pseudo-range observation value based on the receiver clock error in the carrier phase and the pseudo-range observation value; and calculating a standard deviation based on the residual. The calculation process of the residual includes:

[0106] .

[0107] In the formula, R is the residual, P is the pseudo-range observation value, c is the speed of light, is the receiver clock error, and the smaller the value of the residual is, the higher the data quality is. R

[0108] Step 706: calculating a reliability factor through the residual and the standard deviation, including:

[0109] .

[0110] In the formula, F is the reliability factor, r n is the residual of the nth observation value, n is the standard deviation of the residual. The smaller the value of the reliability factor is, the more reliable the data point is.

[0111] Step 707: obtaining a final positioning error based on the position solution accuracy.

[0112] .

[0113] In the formula, e represents the final positioning error, x represents a horizontal axis coordinate obtained by solution, y represents a vertical axis coordinate obtained by solution, z represents a vertical axis coordinate obtained by solution, x true represents a true horizontal axis coordinate, y true represents a true vertical axis coordinate, z true represents a true vertical axis coordinate.

[0114] ​​Step 707 is used to verify whether the progress of the positioning result meets the requirements.

[0115] Step 708: Quality control of the repaired carrier phase observation value by using the residual, reliability factor and final positioning error.

[0116] Figures 2 to 4 The curve schematic comparison diagram provided by the embodiments of the present application. Figure 2 The carrier phase observation value change curve diagram shows the curve comparison of the carrier phase observation value under normal conditions and under the condition of cycle slip. In the Figure 2 , Carrier Phase Change Curve (Normal vs Cycle Slip) means carrier phase observation value change curve (normal condition and cycle slip condition) in Chinese, the horizontal axis Time (Epoch Index) means time (epoch), and the vertical axis Carrier Phase Change means carrier phase in Chinese. Normal Phase Change means normal phase change amount, and Phase Change with Cycle Slips means phase change amount when cycle slip occurs. Cycle Slip means cycle slip in Chinese.

[0117] Figure 3 The cycle slip detection schematic diagram shows the detection effect under different thresholds. In the Figure 3 , Cycle Slip Detection under Different Thresholds means cycle slip detection under different thresholds in Chinese. The horizontal axis Time (Epoch Index) means time, i.e. epoch in Chinese. The vertical axis Carrier Phase Change means carrier phase change amount in Chinese. Carrier Phase Change (with Cycle Slips) means carrier phase change amount when cycle slip occurs. Low Threshold (High False Alarm) means low threshold, which has the characteristics of high alarm sensitivity. Optional Threshold (Best Detection) means threshold of the best detection point. High Threshold (High Miss Rate) means high threshold, which has the characteristics of high miss rate. Cycle Slip Point means cycle slip.

[0118] Figure 4For the comparison curve before and after cycle slip repair, it is mainly used to show the smoothing effect of Kalman filtering. In Figure 4 Cycle Slip Correction: Before and After Kalman Filtering Before Correction (With Cycle Slips) indicates the effect of cycle slip before smoothing by Kalman filtering. After Correction (Kalman Filtering) indicates the effect of cycle slip after smoothing by Kalman filtering. Cycle Slip indicates cycle slip.

[0119] From the above description of the embodiments, those skilled in the art can clearly understand that the present application can be implemented by means of software and necessary hardware. Based on such understanding, the technical solutions of the present application can be embodied in the form of a software product or through the implementation process of data migration. The computer software product can be stored in a storage medium, such as ROM / RAM, magnetic disk, optical disk, etc., and includes a plurality of instructions for causing a computer device (which can be a personal computer, mobile terminal, server, or network device, etc.) to execute the method described in each embodiment or some part of the embodiments of the present application.

[0120] Each embodiment in the specification is described in a progressive manner, and the same or similar parts between each embodiment can be referred to each other, and each embodiment mainly describes the difference from other embodiments. The whole or part of the present application can be used in many general or special computer system environments or configurations. For example: personal computer, server computer, handheld device or portable device, tablet device, mobile communication terminal, multi-processor system, microprocessor-based system, programmable electronic device, network PC, small computer, large computer, distributed computing environment including any of the above systems or devices, etc.

[0121] The above embodiments are only used to illustrate the technical solutions of the present application, and not to limit the present application; although the present application has been described in detail with reference to the foregoing embodiments, those skilled in the art should understand that they can still modify the technical solutions recorded in the foregoing embodiments, or make equivalent replacement for part or all of the technical features; and these modifications or replacements do not make the essence of the corresponding technical solutions deviate from the scope of the technical solutions of the present application.

Claims

1. A method for real-time detection and repair of carrier phase cycle slip in an airborne GNSS receiver, characterized in that, include: Real-time acquisition of GNSS signal data, including carrier phase observations; Based on carrier phase observations, the initial phase change between adjacent epochs is determined; A cycle slip detection model is constructed, and the initial phase change is used as an input parameter to perform preliminary cycle slip detection. The carrier phase observations are smoothed, denoised, and enhanced, and the errors in the carrier phase observations are corrected to obtain corrected carrier phase observations. Based on the corrected carrier phase observations, the corrected phase change is determined. The corrected phase change is used as an input parameter and input into the cycle slip detection model to obtain the initial cycle slip data. The initial cycle slip data includes the satellite in which the cycle slip occurred and the epoch in which the cycle slip occurred; Repair the initial cycle slip data to obtain cycle slip data, and obtain the repaired carrier phase observations based on the cycle slip data; The baseline solution results are re-determined based on the repaired carrier phase observations, and the quality control of the repaired carrier phase observations is performed. Determining the initial phase change between adjacent epochs includes: ； In the formula, This represents the initial phase change between adjacent epochs. Indicates the first k Carrier phase observations at each epoch. Indicates the first k Carrier phase observations at epoch -1; The preliminary cycle slip detection includes: Set a cycle slip threshold and calculate the mean and standard deviation of the initial phase change. Based on the initial phase change, mean, standard deviation, and cycle slip threshold, the epoch range in which cycle slips occur is preliminarily determined; The abrupt change point is obtained based on wavelet transform; the abrupt change point is used to verify or supplement the epoch range of cycle slips. The process of repairing the initial cycle slip data to obtain cycle slip data includes: Based on regression-based prediction methods, the initial cycle slip data is repaired, including: The missing values ​​in the initial cycle slip data are filled in using linear interpolation to obtain the filled cycle slip data; A dynamic repair model is established based on the recursive least squares algorithm. The cycle slip data is used as input parameters to the dynamic repair model to optimize the repair results and obtain cycle slip data. After obtaining the cycle slip data, the following is also included: The filled-in cycle slip data and cycle slip data are used as training samples and input into the long short-term memory network learning model for training; Based on the trained Long Short-Term Memory network learning model, the cycle slip data in other epochs is predicted and repaired to obtain the predicted and repaired cycle slip data.

2. The method according to claim 1, characterized in that, The verification or supplementation of the epoch range of cycle slips based on wavelet transform includes: The abrupt change characteristics of the initial phase change are calculated based on wavelet transform to obtain the abrupt change point; Based on the mutation point, verify or supplement the epoch range where cycle slips occur.

3. The method according to claim 1, characterized in that, The smoothing, denoising, and enhancement processing of the carrier phase observations includes: Multidimensional filtering methods are used to smooth, denoise, and enhance carrier phase observations, including: The current state of the carrier phase observations is predicted using the conventional Kalman filter method, and the Kalman gain of the carrier phase observations is calculated. Smooth the carrier phase observations based on the current state and Kalman gain; The smoothed carrier phase observations are denoised using median filtering. We used Weiner filtering to enhance the signal of the denoised carrier phase observations.

4. The method according to claim 1, characterized in that, The error of the corrected carrier phase observation is obtained; the corrected carrier phase observation includes: Calculation based on carrier phase observation equation i The satellite in the k Carrier phase of an epoch; Calculate the first based on the pseudorange observation equation i The satellite in the k Pseudorange observations at each epoch; The error range of the carrier phase is calculated using the error propagation formula; The error range is corrected by using pseudorange observations to obtain the corrected carrier phase observations.

5. The method according to claim 1, characterized in that, The recalculation of the baseline solution based on the repaired carrier phase observations includes: The carrier phase of the repaired carrier phase observation is determined based on the carrier phase observation equation, and the pseudorange observation value of the repaired carrier phase observation is determined based on the pseudorange observation equation. Based on carrier phase and pseudorange observations, a double-difference observation equation is constructed, and common error terms are eliminated; the common error terms include receiver clock bias and satellite clock bias. Based on the least squares method, the ambiguity parameters in the double-difference observation equation are estimated. Based on the ambiguity parameters and the double-difference observation equation, the baseline solution results are determined. The baseline solution results include the baseline vector and the covariance matrix of the baseline vector. The baseline vector represents the relative positional relationship between the two airborne GNSS receivers, and the covariance matrix represents the accuracy of the baseline solution results.

6. The method according to claim 5, characterized in that, The quality control of the repaired carrier phase observations includes: Based on the receiver clock error and pseudorange observations in the repaired carrier phase observations, the residuals of the pseudorange observations are obtained; Determine the standard deviation based on the residuals of pseudorange observations; The reliability factor is calculated using the residuals and standard deviations of pseudorange observations. The final positioning error is obtained based on the accuracy of the position calculation. The residuals, reliability factors, and final positioning errors are fed back to the repaired carrier phase observations to control the quality of the repaired carrier phase observations.

Images