A method and system for predicting precision parameters of time-varying messages from navigation satellites

The long-term memory neural network algorithm calculates the prediction value of satellite orbit and clock error of the GNSS system, which solves the contradiction between "packet stay" and "packet tightness" when predicting the accuracy of satellite ephemeris and clock difference, and improves the The envelope capability of broadcast ephemeris accuracy parameters improves the service accuracy and availability of the GNSS system.

CN118377041BActive Publication Date: 2024-11-26SHANGHAI ASTRONOMICAL OBSERVATORY CHINESE ACAD OF SCI +1
View PDF 0 Cites 0 Cited by

Patent Information

Application Number
CN202410480115.9
Authority / Receiving Office
CN · China
Patent Type
Patents(China)
Current Assignee / Owner
Filing Date
2024-04-19
Publication Date
2024-11-26
Estimated Expiration
2044-04-19

AI Technical Summary

Technical Problem

When predicting the accuracy of the satellite ephemeris and clock difference, there is a contradiction between "package stay" and "package tightness" in the existing GNSS system, which makes it difficult to accurately reflect the actual service accuracy of the system when the broadcast value of the broadcast ephemeris accuracy parameter is too large. Causes reduced service availability.

Method used

The long and short-term memory neural network algorithm is used to calculate orbit error and clock error based on the broadcast ephemeris and post-precision ephemeris of the satellite navigation system, generate orbit error prediction values ​​and clock difference prediction values, and calculate orbit accuracy parameters and clock difference accuracy parameters. And convert it into accuracy parameter level values ​​to improve the envelope capability of broadcast ephemeris accuracy parameters.

Benefits of technology

It effectively improves the envelope capability of broadcast ephemeris accuracy parameters to spatial signal errors, improves the integrity of the GNSS system, and ensures the service accuracy and availability of the system.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure CN118377041B_ABST
    Figure CN118377041B_ABST
Patent Text Reader

Abstract

The present invention relates to the field of satellite navigation technology, and proposes a method and system for predicting the accuracy parameters of time-varying telegrams of navigation satellites. The method includes the following steps: calculating the broadcast ephemeris orbit error and the broadcast ephemeris clock error according to the broadcast ephemeris and post-precision ephemeris of the satellite navigation system; generating a broadcast ephemeris orbit error prediction value and a broadcast ephemeris clock error prediction value through a long short-term memory neural network algorithm according to the broadcast ephemeris orbit error and the broadcast ephemeris clock error; calculating the broadcast ephemeris orbit accuracy parameters and the broadcast ephemeris clock error accuracy parameters according to the broadcast ephemeris orbit error prediction value and the broadcast ephemeris clock error prediction value; and converting the broadcast ephemeris orbit accuracy parameters and the broadcast ephemeris clock error accuracy parameters into broadcast ephemeris accuracy parameter level values. The present invention can effectively improve the envelope capability of the broadcast ephemeris accuracy parameters to the broadcast ephemeris space signal error, thereby improving the integrity service capability of the broadcast ephemeris accuracy parameters.
Need to check novelty before this filing date? Find Prior Art

Description

Technical Field

[0001] The present invention generally relates to the field of satellite navigation technology. Specifically, the present invention relates to a method and system for predicting time-varying message accuracy parameters of a navigation satellite. Background Art

[0002] The Global Navigation Satellite System (GNSS) refers to a satellite navigation system that can provide accurate, real-time and global spatiotemporal information to near-Earth users by broadcasting broadcast ephemeris information. Integrity is one of the important indicators of the GNSS system. It is used to characterize the ability to promptly warn users when an abnormality occurs in the system and causes service performance to exceed the limit. It is a key item to determine whether the system is safe and reliable. In order to ensure the integrity service capability, each GNSS system broadcasts the accuracy parameters of the broadcast ephemeris in the broadcast ephemeris and predicts the accuracy of the satellite orbit and clock error broadcast by the broadcast ephemeris. Through the broadcast ephemeris accuracy parameters, users are informed of the accuracy alarm threshold of the orbit and clock error information broadcast by the GNSS satellite. If the signal in space range error (SISRE) of the broadcast ephemeris exceeds the broadcast ephemeris accuracy parameter, a system integrity risk is generated. However, when the broadcast value of the broadcast ephemeris accuracy parameter is too large, it will be difficult to accurately represent the actual service accuracy of the system, resulting in reduced service availability.

[0003] Due to the complex space-air environment, it is difficult to accurately predict the accuracy of satellite ephemeris and clock errors. There is a contradiction between the broadcast ephemeris accuracy parameters and the space signal ranging error (SISRE). At present, in order to ensure the integrity of the system service capability, various GNSS systems usually broadcast fixed parameters that are about several dozen times greater than SISRE as the message accuracy threshold based on the long-term historical data statistics of SISRE of the broadcast ephemeris. This cannot objectively reflect the actual service level of the system, and can only solve the problem of the broadcast ephemeris accuracy parameters "encompassing" SISRE, but cannot solve the problem of "encompassing tightly". Summary of the invention

[0004] In order to at least partially solve the above problems in the prior art, the present invention proposes a method for predicting accuracy parameters of a navigation satellite time-varying message, comprising the following steps:

[0005] Calculate the broadcast ephemeris orbit error and broadcast ephemeris clock error based on the broadcast ephemeris and post-precision ephemeris of the satellite navigation system;

[0006] Generate a broadcast ephemeris orbit error prediction value and a broadcast ephemeris clock error prediction value through a long short-term memory neural network algorithm according to the broadcast ephemeris orbit error and the broadcast ephemeris clock error;

[0007] Calculating a broadcast ephemeris orbit accuracy parameter and a broadcast ephemeris clock error accuracy parameter according to the broadcast ephemeris orbit error prediction value and the broadcast ephemeris clock error prediction value; and

[0008] The broadcast ephemeris orbit accuracy parameter and the broadcast ephemeris clock error accuracy parameter are converted into a broadcast ephemeris accuracy parameter level value.

[0009] In one embodiment of the present invention, it is provided that calculating the broadcast ephemeris orbit error includes performing a satellite antenna phase center deviation correction; and / or

[0010] Calculation of broadcast ephemeris clock error includes correction of satellite antenna phase center deviation, hardware delay correction and clock reference deviation correction.

[0011] In one embodiment of the present invention, generating a broadcast ephemeris orbit error prediction value and a broadcast ephemeris clock error prediction value by using a long short-term memory neural network algorithm includes the following steps:

[0012] Set the training sample duration, and select the broadcast ephemeris orbit error time series and broadcast ephemeris clock error time series as training samples according to the training sample duration;

[0013] Normalizing the broadcast ephemeris orbit error time series and the broadcast ephemeris clock error time series;

[0014] Construct a long short-term memory neural network model;

[0015] Setting the predicted arc length by the long short-term memory neural network model, training and predicting the broadcast ephemeris orbit error time series and the broadcast ephemeris clock error time series; and

[0016] The prediction results are denormalized to generate broadcast ephemeris orbit error prediction values ​​and broadcast ephemeris clock error prediction values.

[0017] In one embodiment of the present invention, it is provided that a broadcast ephemeris orbit error prediction value and a broadcast ephemeris clock error prediction value CLK of each epoch are obtained according to a sampling interval, wherein the broadcast ephemeris orbit error prediction value includes an orbit radial direction error prediction value R, an orbit normal direction error prediction value N, and an orbit tangential direction error prediction value T.

[0018] In one embodiment of the present invention, it is provided that the broadcast ephemeris orbit error prediction value and the broadcast ephemeris clock error prediction value are obtained with a broadcast ephemeris update period of 1 hour and a sampling interval of 5 minutes, and 12 groups of orbit radial direction error prediction values ​​R, orbit normal direction error prediction values ​​N, orbit tangential direction error prediction values ​​and broadcast ephemeris clock error prediction values ​​CLK are obtained every hour, and the satellite orbit and clock error prediction error prediction values ​​Δi It is expressed as the following formula:

[0019]

[0020] Here, i represents hours, i=1, 2, ..., 24.

[0021] In one embodiment of the present invention, it is provided that calculating the broadcast ephemeris orbit accuracy parameter includes:

[0022] The broadcast ephemeris orbit accuracy parameters are calculated based on the orbit tangent direction error prediction value T and the orbit normal direction error prediction value N. One broadcast ephemeris orbit accuracy parameter is calculated every hour, where the orbit tangent plane prediction accuracy SISA is calculated for each epoch within one hour. oe,i,j , and take the orbit tangent plane prediction accuracy within one hour SISA oe,i,j The maximum value is taken as the broadcast ephemeris orbit accuracy parameter SISA for that hour oe,i , expressed as the following formula:

[0023]

[0024] Where i represents hour, j represents each sampling point within an hour, j = 1, 2, ..., 12.

[0025] In one embodiment of the present invention, it is provided that calculating the broadcast ephemeris clock error accuracy parameter includes:

[0026] Calculate the broadcast ephemeris clock error accuracy parameter RCLK for each error sampling point based on the orbit radial direction error prediction value R and the broadcast ephemeris clock error prediction value CLK i,j , expressed as the following formula:

[0027] RCLK i,j =|R i,j |+|CLK i,j |

[0028] Where i represents hour, j represents each sampling point within an hour, j = 1, 2, ..., 12.

[0029] In one embodiment of the present invention, it is provided that calculating the broadcast ephemeris clock error accuracy parameter includes:

[0030] Calculate the broadcast ephemeris clock error accuracy parameters of multiple sets of broadcast ephemeris, where the ground uploads one day of broadcast ephemeris to the satellite at a time, the update cycle of the broadcast ephemeris is one hour, and a total of 24 sets of broadcast ephemeris are uploaded;

[0031] When calculating the broadcast ephemeris clock error accuracy parameters of each set of broadcast ephemeris, the clock error prediction error time series within one hour starting from the reference time t0 of the set of broadcast ephemeris is selected for fitting;

[0032] If the reference time t0 is less than 26 hours from the current time, a first-order linear model is used to fit the constant term parameter SISA′ of the broadcast ephemeris clock error accuracy parameter. ocb,i and speed parameter SISA oc1i If the reference time t0 is greater than or equal to 26 hours from the current time, a second-order linear model is used to fit the constant term parameter SISA′ of the broadcast ephemeris clock error accuracy parameter ocb,i , speed parameter SISA oc1,i And the acceleration parameter SISA oc2,i , expressed as the following formula:

[0033] RCLK i,j =SISA' ocb,i +SISA oc1,i *(t j -t0) tt 0,i ≤26h

[0034] RCLK i,j =SISA' ocb,i +SISA ocl,i *(t-t0)+SISA oc2,i *(t j -t0) 2 tt 0,i >26h

[0035] Where i represents the i-th set of broadcast ephemeris, t represents the current calculation time, and t 0,i represents the reference time of the i-th set of broadcast ephemeris, t j represents the time of each clock error sampling point in the clock error prediction error time series used in calculating the i-th set of broadcast ephemeris;

[0036] Calculate the mean square error information rms (RLCK) of the clock error prediction time series used in the i-th group of broadcast ephemeris i , expressed as the following formula:

[0037] as well as

[0038] Select the maximum value of the bias term and the mean square error information of the clock error time series in the fitting model as the bias term SISA of the i-th group of broadcast ephemeris clock error accuracy parameters ocb,i , expressed as the following formula:

[0039] SISA ocb,i =max[SISA′ ocb,i ,rms(RCLK) i ].

[0040] In one embodiment of the present invention, the broadcast ephemeris accuracy parameter level value includes the broadcast ephemeris orbit accuracy parameter level value SISAI oe , broadcast ephemeris clock error accuracy constant parameter level value SISAI ocb , broadcast ephemeris clock error accuracy velocity item parameter level value SISAI oc1 And the broadcast ephemeris clock error accuracy acceleration parameter level value SISAI oc2 , where the broadcast ephemeris orbit accuracy parameter level value SISAI oe , broadcast ephemeris clock error accuracy constant parameter level value SISAI ocb Take the upper limit value in the index relationship table, the broadcast ephemeris clock error accuracy speed item parameter level value SISAI oc1 And the broadcast ephemeris clock error accuracy acceleration parameter level value SISAI oc2 The transformation is expressed as follows:

[0041] SISAI oc1 =-log2SISA oc1 -14

[0042] SISAI oc2 =-log2SISA oc1 -28.

[0043] The present invention also proposes a navigation satellite time-varying message accuracy parameter prediction system, comprising:

[0044] A broadcast ephemeris error calculation module, which is configured to calculate the broadcast ephemeris orbit error and the broadcast ephemeris clock error according to the broadcast ephemeris and post-precision ephemeris of the satellite navigation system;

[0045] A broadcast ephemeris error prediction module, configured to generate a broadcast ephemeris orbit error prediction value and a broadcast ephemeris clock error prediction value through a long short-term memory neural network algorithm according to the broadcast ephemeris orbit error and the broadcast ephemeris clock error;

[0046] a broadcast ephemeris accuracy parameter calculation module, configured to calculate a broadcast ephemeris orbit accuracy parameter and a broadcast ephemeris clock error accuracy parameter according to the broadcast ephemeris orbit error prediction value and the broadcast ephemeris clock error prediction value; and

[0047] The broadcast ephemeris accuracy parameter level value conversion module is configured to convert the broadcast ephemeris orbit accuracy parameter and the broadcast ephemeris clock error accuracy parameter into a broadcast ephemeris accuracy parameter level value.

[0048] The present invention has at least the following beneficial effects: the present invention can effectively improve the envelope capability of the broadcast ephemeris precision parameter to the broadcast ephemeris space signal error, thereby improving the integrity service capability of the broadcast ephemeris precision parameter. The reason is that when the broadcast ephemeris is updated, the telegram accuracy may jump due to the change of the information processing data source, and the broadcast ephemeris precision parameter actually broadcast by the system cannot predict the related jump. When the telegram accuracy jumps, there is a risk of missing the envelope. The broadcast ephemeris precision parameter calculated by the present invention can accurately predict the accuracy jump caused by the telegram update, so the envelope capability of the calculated broadcast ephemeris precision parameter to the broadcast ephemeris space signal error can reach 100%. BRIEF DESCRIPTION OF THE DRAWINGS

[0049] To further illustrate the advantages and features of various embodiments of the present invention, a more detailed description of various embodiments of the present invention will be presented with reference to the accompanying drawings. It will be understood that these drawings only depict typical embodiments of the present invention and are therefore not to be considered as limiting the scope thereof. In the accompanying drawings, for clarity, identical or corresponding parts will be represented by identical or similar reference numerals.

[0050] Figure 1 The present invention shows a flowchart of a method for predicting time-varying message accuracy parameters of a navigation satellite in one embodiment of the present invention.

[0051] Figure 2 A schematic diagram showing a comparison between a track prediction result and a post-evaluation result in one embodiment of the present invention is shown.

[0052] Figure 3 A schematic diagram showing a comparison between a clock error prediction result and a post-evaluation result in one embodiment of the present invention is shown.

[0053] Figure 4 A schematic diagram showing the envelope capability of the actual ephemeris accuracy parameter SISA broadcast by the BeiDou satellite navigation system to the space signal error.

[0054] Figure 5 A schematic diagram showing the envelope capability of the ephemeris accuracy parameter error calculated in one embodiment of the present invention.

[0055] Figure 6 A schematic diagram showing the envelope performance of the broadcast ephemeris accuracy parameters and broadcast ephemeris errors of the C21, C33 and C41 satellites in the Beidou satellite navigation system.

[0056] Figure 7 A schematic diagram showing the comparison of the envelope margin of the orbit accuracy parameter to the spatial signal error in one embodiment of the present invention is shown. DETAILED DESCRIPTION

[0057] It should be noted that the components in the figures may be shown exaggeratedly for the sake of illustration and are not necessarily true to scale. In the figures, identical or functionally identical components are provided with the same reference numerals.

[0058] In the present invention, unless otherwise specified, "arranged on...", "arranged above..." and "arranged above..." do not exclude the existence of an intermediate between the two. In addition, "arranged on or above..." merely indicates the relative positional relationship between two components, and in certain cases, such as after reversing the product direction, it can also be converted into "arranged below or below...", and vice versa.

[0059] In the present invention, each embodiment is only intended to illustrate the aspects of the present invention and should not be construed as limiting.

[0060] In the present invention, unless otherwise specified, the quantifiers "a", "an" and "an" do not exclude the presence of a plurality of elements.

[0061] It should also be noted that in the embodiments of the present invention, for the sake of clarity and simplicity, only a portion of the parts or components may be shown, but those of ordinary skill in the art will understand that under the teachings of the present invention, the required parts or components may be added according to the needs of the specific scenario. In addition, unless otherwise specified, the features in different embodiments of the present invention may be combined with each other. For example, a feature in the second embodiment may be used to replace a corresponding or functionally identical or similar feature in the first embodiment, and the resulting embodiment also falls within the disclosure scope or recorded scope of the present application.

[0062] It should also be noted that within the scope of the present invention, the terms "same", "equal", "equal to" and the like do not mean that the values ​​of the two are absolutely equal, but allow a certain reasonable error, that is, the terms also cover "substantially the same", "substantially equal", "substantially equal to". By analogy, in the present invention, the terms "perpendicular to", "parallel to" and the like indicating directions also cover the meanings of "substantially perpendicular to" and "substantially parallel to".

[0063] In addition, the numbering of the steps of the methods of the present invention does not limit the execution order of the method steps. Unless otherwise specified, the method steps can be executed in different orders.

[0064] The present invention will be further described below in conjunction with specific embodiments with reference to the accompanying drawings.

[0065] The present invention provides a method for predicting the accuracy parameters of a navigation satellite time-varying message, comprising the following steps:

[0066] Calculate the broadcast ephemeris orbit error and broadcast ephemeris clock error based on the broadcast ephemeris and post-precision ephemeris of the satellite navigation system;

[0067] Generate a broadcast ephemeris orbit error prediction value and a broadcast ephemeris clock error prediction value through a long short-term memory neural network algorithm according to the broadcast ephemeris orbit error and the broadcast ephemeris clock error;

[0068] Calculating a broadcast ephemeris orbit accuracy parameter and a broadcast ephemeris clock error accuracy parameter according to the broadcast ephemeris orbit error prediction value and the broadcast ephemeris clock error prediction value; and

[0069] The broadcast ephemeris orbit accuracy parameter and the broadcast ephemeris clock error accuracy parameter are converted into a broadcast ephemeris accuracy parameter level value.

[0070] Figure 1 The following is a schematic diagram showing a process flow of a method for predicting time-varying message accuracy parameters of a navigation satellite in one embodiment of the present invention. Figure 1 The steps of this method are described in detail:

[0071] Step 1: Use the subsequent precise ephemeris and broadcast ephemeris to calculate the broadcast ephemeris orbit error and clock error for the past period of time based on the precise ephemeris. The calculation of the broadcast ephemeris orbit error includes correcting the satellite's antenna phase center deviation, and the calculation of the broadcast ephemeris clock error includes correcting the satellite antenna phase center deviation, hardware delay correction, and clock reference deviation correction.

[0072] Step 2: Set the training sample duration. According to the training duration configuration, select the most recent broadcast ephemeris orbit error time series and clock error time series as the training set.

[0073] Step 3: Normalize the orbit error and clock error so that the orbit error and clock error range is [-1, 1] to speed up the fitting of subsequent models.

[0074] Step 4: Build a long short-term memory neural network algorithm (LSTM) training model and set the key parameters of the LSTM training model, including an initial learning rate of 0.005, 200 hidden layer neurons, and a maximum number of training iterations of 150.

[0075] Step 5: Use the LSTM training model to set the predicted arc length, train and predict the broadcast ephemeris orbit error training samples. Similarly, train and predict the broadcast ephemeris clock error training samples.

[0076] Step 6: Denormalize the prediction results to obtain the predicted values ​​of the broadcast ephemeris orbit error and the broadcast ephemeris clock error within the predicted arc length period.

[0077] The predicted value of the broadcast ephemeris orbit error for each epoch is obtained at a sampling interval of 5 minutes. The predicted value of the broadcast ephemeris orbit error includes the results in the three directions of orbit radial (R), normal (N) and tangential (T). And the predicted value of the broadcast ephemeris clock error (CLK) for each epoch is obtained at a sampling interval of 5 minutes. Usually, the broadcast ephemeris update cycle is 1 hour. Taking the 1-hour update cycle as an example, 12 sets of satellite orbit R, T, N direction prediction errors and satellite clock error prediction errors can be obtained every hour. The satellite orbit and clock error prediction errors for 24 hours a day can be calculated as △ i It is expressed as the following formula:

[0078]

[0079] Wherein i represents hour, i=1,2,...,24.

[0080] Step 7: Calculate the orbital accuracy parameters of the broadcast ephemeris. The satellite orbit prediction accuracy is expressed by the combined error in the tangent and normal directions of the satellite orbit. Calculate a broadcast ephemeris orbital accuracy parameter every hour. First, calculate the orbital tangent plane prediction accuracy SISA for each epoch in one hour. oe,i,j , take SISA within one hour oe,i,j The maximum value is taken as the satellite orbit prediction accuracy SISA for that hour oe,i , expressed as the following formula:

[0081]

[0082] Where i represents hour, j = 1, 2, ..., 12, which are the sampling points within an hour.

[0083] Step 8: Calculate the broadcast ephemeris clock error accuracy parameters. The satellite clock error prediction accuracy is expressed by the combined error of the satellite orbit radial error and the satellite clock error. For each error sampling point, the orbit radial and clock error combined prediction error can be expressed as follows:

[0084] RCLK i,j =|R i,j |+|CLK i,j |

[0085] Where i represents hour, and j = 1, 2, ..., 12 represents each sampling point within an hour.

[0086] When the ground is notifying the satellite of broadcast ephemeris, usually one day of broadcast ephemeris is notified at a time, and the broadcast ephemeris update cycle is 1 hour, so 24 sets of broadcast ephemeris need to be notified. When calculating the clock error parameter accuracy of each set of broadcast ephemeris, the clock error prediction error time series (12 sampling points) within one hour starting from the reference time t0 of the broadcast ephemeris of this set is selected, and different fitting models are selected according to the prediction time length between the reference time t0 of the broadcast ephemeris and the current time.

[0087] Specifically, if the forecast broadcast ephemeris reference time is less than 26 hours from the current time, the first-order linear model is used to fit the clock error accuracy parameters. If the forecast broadcast ephemeris reference time is greater than or equal to 26 hours from the current time, the second-order linear model is used to fit the clock error accuracy parameters, which is expressed as the following formula:

[0088] RCLK i,j =SISA' ocb,i +SISA oc1,i *(t j -t0) tt 0,i ≤26h

[0089] RCLK i,j =SISA' ocb,i +SISA oc1,i *(t-t0)+SISA oc2,i *(t j -t0) 2 tt 0,i >26h

[0090] Where i represents the i-th set of broadcast ephemeris, t represents the current calculation time, and t 0,i represents the reference time of the i-th set of broadcast ephemeris, t j It represents the time of each clock error sampling point in the clock error prediction error time series used in calculating the i-th set of broadcast ephemeris.

[0091] And calculate the mean square error information of the clock error prediction time series used by the i-th group of broadcast ephemeris, which is expressed as follows:

[0092]

[0093] In order to ensure the envelope capability of the clock error accuracy parameter to the broadcast ephemeris clock error, the bias term SISA of the i-th group of broadcast ephemeris clock error accuracy parameters is ocb,i Select the maximum value of the bias term and the root mean square information of the clock error time series in the fitting model, which is expressed as follows:

[0094] SISA ocb,i =max[SISA′ ocb,i ,rms(RCLK) i ].

[0095] Step 9: Calculate the orbit accuracy parameter level value and clock error accuracy parameter level conversion value, where SISAI oe and SISAI ocb Take the upper limit value in the corresponding index relationship table respectively. oc1 and SISAI oc2 The transformation is expressed as follows:

[0096] SISAI oc1 =-log2SISA oc1 -14

[0097] SISAI oc2 =-log2SISA oc1 -28.

[0098] SISAI oe The index range is +15 to -16, which is consistent with the SISA oe The relationship is shown in Table 1. ocb The index range is +15 to -16, which is consistent with the SISA ocb The relationship is shown in Table 2.

[0099] Table 1

[0100]

[0101]

[0102] Table 2

[0103]

[0104]

[0105] The present invention also proposes a navigation satellite time-varying message accuracy parameter prediction system, comprising:

[0106] A broadcast ephemeris error calculation module, which is configured to calculate the broadcast ephemeris orbit error and the broadcast ephemeris clock error according to the broadcast ephemeris and post-precision ephemeris of the satellite navigation system;

[0107] A broadcast ephemeris error prediction module, configured to generate a broadcast ephemeris orbit error prediction value and a broadcast ephemeris clock error prediction value through a long short-term memory neural network algorithm according to the broadcast ephemeris orbit error and the broadcast ephemeris clock error;

[0108] a broadcast ephemeris accuracy parameter calculation module, configured to calculate a broadcast ephemeris orbit accuracy parameter and a broadcast ephemeris clock error accuracy parameter according to the broadcast ephemeris orbit error prediction value and the broadcast ephemeris clock error prediction value; and

[0109] The broadcast ephemeris accuracy parameter level value conversion module is configured to convert the broadcast ephemeris orbit accuracy parameter and the broadcast ephemeris clock error accuracy parameter into a broadcast ephemeris accuracy parameter level value.

[0110] Figure 2 FIG. 2 shows a schematic diagram of comparing the orbit prediction result and the post-evaluation result in one embodiment of the present invention. Figure 2 As shown in the figure, the 24-hour orbit prediction results based on the LSTM prediction algorithm are compared with the post-evaluation results. The prediction accuracy of the RTN orbit direction is less than 0.05 meters. Figure 3 FIG. 1 is a schematic diagram showing a comparison between a clock error prediction result and a post-evaluation result in one embodiment of the present invention. Figure 3 As shown, the 24-hour clock error prediction result based on the LSTM prediction algorithm is compared with the post-evaluation result, and the clock error prediction accuracy is less than 0.3 meters.

[0111] In order to illustrate the envelope capability of the broadcast ephemeris accuracy parameters to the broadcast ephemeris space signal error, that is, the integrity service capability of the broadcast ephemeris accuracy parameters, the SISA parameters actually broadcast in the satellite broadcast ephemeris of the Beidou Satellite Navigation System (BDS 3) and the SISA parameters calculated by the present invention are selected to calculate the envelope capability of SISAoe to the orbit error, the envelope capability of SISAoc to the orbit radial and clock error, and the envelope capability of the SISA parameters combined by SISAoe and SISAoc to the worst user direction SISRE error.

[0112] Figure 4 A schematic diagram showing the envelope capability of the actual ephemeris accuracy parameter SISA broadcast by the BeiDou satellite navigation system to the space signal error. Figure 5 FIG. 1 is a schematic diagram showing the envelope capability of the ephemeris accuracy parameter error calculated in one embodiment of the present invention. Figure 4 and Figure 5 As shown, the envelope capability of the broadcast ephemeris precision parameters of the actual broadcast ephemeris of BDS-3 to the space signal error is about 99.9%, and the envelope capability of the SISAoe of all BDS-3 satellites calculated by the present invention to the orbit error, the envelope capability of the SISAoc to the orbit radial and clock error, and the envelope capability of the comprehensive SISA parameters to the SISRE in the worst user direction all reach 100%. It can be seen that compared with the broadcast ephemeris precision parameters actually broadcast by the broadcast ephemeris, the use of this method can significantly improve the integrity service capability of the broadcast ephemeris precision parameters.

[0113] The reason why the present invention can improve the integrity service capability of the broadcast ephemeris precision parameters is that when the broadcast ephemeris is updated, the accuracy jump of the message may occur due to the change of the information processing data source, and the broadcast ephemeris precision parameters actually broadcast by the system cannot predict the relevant jump. When the message accuracy jumps, there is a risk of missing the envelope. The broadcast ephemeris precision parameters calculated by the present invention can accurately predict the accuracy jump caused by the message update, so the envelope capability of the calculated broadcast ephemeris precision parameters for the broadcast ephemeris space signal error can reach 100%. Figure 6The figure shows the envelope performance diagram of the broadcast ephemeris accuracy parameters and broadcast ephemeris error of the C21, C33 and C41 satellites in the Beidou satellite navigation system. Figure 6 As shown, the red curve represents the broadcast ephemeris space signal error, the black curve in the left figure represents the broadcast ephemeris accuracy parameters actually broadcast by the Beidou satellite navigation system, and the black curve in the right figure represents the broadcast ephemeris accuracy parameters calculated by the present invention.

[0114] The present invention not only accurately predicts the accuracy of broadcast message orbit and clock error parameters, but also significantly reduces the envelope margin of orbit accuracy parameter SISAoe to SISRE, solves the problem of "tight package", and can objectively reflect the real service accuracy of the system and improve the system service commitment capability. Taking the C21, C33 and C41 satellites of the Beidou satellite navigation system as an example, the envelope tightness of the orbit accuracy parameters broadcast by the system broadcast ephemeris and the orbit accuracy parameters calculated by the present invention to the space signal error is compared, as shown in Table 3.

[0115] Table 3

[0116]

[0117]

[0118] Figure 7 FIG. 2 shows a schematic diagram of the comparison of the orbit accuracy parameter and the envelope margin of the spatial signal error in one embodiment of the present invention. Figure 7 As shown, taking the C33 satellite as an example, the broadcast ephemeris orbit accuracy parameters (black) actually broadcast by the BDS-3 system and the broadcast ephemeris orbit accuracy parameters (blue) calculated by the present invention are shown to show the envelope performance of the broadcast ephemeris space signal error. The results show that the envelope tightness of the orbit accuracy parameters calculated by the present invention to the space signal error is significantly improved.

[0119] Although various embodiments of the present invention are described above, it should be understood that they are presented as examples only and not as limitations. It is obvious to those skilled in the relevant art that various combinations, modifications and changes can be made thereto without departing from the spirit and scope of the present invention. Therefore, the breadth and scope of the present invention disclosed herein should not be limited by the exemplary embodiments disclosed above, but should only be defined according to the attached claims and their equivalents.

Claims

1. A method for predicting the accuracy parameters of a navigation satellite time-varying message, characterized in that: The following steps are involved: Calculate the broadcast ephemeris orbit error and broadcast ephemeris clock error based on the broadcast ephemeris and post-precision ephemeris of the satellite navigation system; Generate a broadcast ephemeris orbit error prediction value and a broadcast ephemeris clock error prediction value through a long short-term memory neural network algorithm according to the broadcast ephemeris orbit error and the broadcast ephemeris clock error; Calculating a broadcast ephemeris orbit accuracy parameter and a broadcast ephemeris clock error accuracy parameter according to the broadcast ephemeris orbit error prediction value and the broadcast ephemeris clock error prediction value; as well as Converting the broadcast ephemeris orbit accuracy parameter and the broadcast ephemeris clock error accuracy parameter into a broadcast ephemeris accuracy parameter level value; The broadcast ephemeris orbit error prediction value and the broadcast ephemeris clock error prediction value CLK of each epoch are obtained according to the sampling interval, wherein the broadcast ephemeris orbit error prediction value includes an orbit radial direction error prediction value R, an orbit normal direction error prediction value N, and an orbit tangential direction error prediction value T; The orbit accuracy parameters for calculating broadcast ephemeris include: The broadcast ephemeris orbit accuracy parameters are calculated based on the orbit tangent direction error prediction value T and the orbit normal direction error prediction value N. One broadcast ephemeris orbit accuracy parameter is calculated every hour, where the orbit tangent plane prediction accuracy SISA is calculated for each epoch within one hour. oe,i,j , and take the orbit tangent plane prediction accuracy within one hour SISA oe,i,j The maximum value is taken as the broadcast ephemeris orbit accuracy parameter SISA for that hour oe,i , expressed as the following formula: Where i represents the hour, j represents the sampling points within an hour, j = 1, 2, ..., 12; The parameters for calculating the broadcast ephemeris clock error accuracy include: Calculate the broadcast ephemeris clock error accuracy parameter RCLK for each error sampling point based on the orbit radial direction error prediction value R and the broadcast ephemeris clock error prediction value CLK i,j , expressed as the following formula: RCLK i,j =|R i,j |+|CLK i,j | Where i represents the hour, j represents the sampling points within an hour, j = 1, 2, ..., 12; Calculate the broadcast ephemeris clock error accuracy parameters of multiple sets of broadcast ephemeris, where the ground uploads one day of broadcast ephemeris to the satellite at a time, the update cycle of the broadcast ephemeris is one hour, and a total of 24 sets of broadcast ephemeris are uploaded; When calculating the broadcast ephemeris clock error accuracy parameters of each set of broadcast ephemeris, the clock error prediction error time series within one hour starting from the reference time t0 of the set of broadcast ephemeris is selected for fitting; If the reference time t0 is less than 26 hours from the current time, a first-order linear model is used to fit the constant term parameter SISA′ of the broadcast ephemeris clock error accuracy parameter. ocb,i and speed parameter SISA oc1,i If the reference time t0 is greater than or equal to 26 hours from the current time, a second-order linear model is used to fit the constant term parameter SISA′ of the broadcast ephemeris clock error accuracy parameter ocb,i , speed parameter SISA oc1,i And the acceleration parameter SISA oc2,i , expressed as the following formula: RCLK i,j =SISA' ocb,i +SISA oc1,i *(t j -t0) t-t 0,i ≤26h RCLK i,j =SISA' ocb,i +SISA oc1,i *(t-t0)+SISA oc2,i *(t j -t0) 2 t-t 0,i >26h Where i represents the i-th set of broadcast ephemeris, t represents the current calculation time, and t 0,i represents the reference time of the i-th set of broadcast ephemeris, t j represents the time of each clock error sampling point in the clock error prediction error time series used in calculating the i-th set of broadcast ephemeris; Calculate the mean square error information rms (RLCK) of the clock error prediction time series used in the i-th group of broadcast ephemeris i , expressed as the following formula: as well as Select the maximum value of the bias term and the mean square error information of the clock error time series in the fitting model as the bias term SISA of the i-th group of broadcast ephemeris clock error accuracy parameters ocb,i , expressed as the following formula: REMAINDER ocb,i =max[REMAINING' ocb,i ,rms(RCLK) i ]。 2. The method for predicting the accuracy parameters of navigation satellite time-varying messages according to claim 1, characterized in that: Calculating broadcast ephemeris orbit errors includes making satellite antenna phase center deviation corrections; and / or Calculation of broadcast ephemeris clock error includes correction of satellite antenna phase center deviation, hardware delay correction and clock reference deviation correction.

3. The method for predicting the accuracy parameters of navigation satellite time-varying messages according to claim 1, characterized in that: Generating the broadcast ephemeris orbit error prediction value and the broadcast ephemeris clock error prediction value by the long short-term memory neural network algorithm includes the following steps: Set the training sample duration, and select the broadcast ephemeris orbit error time series and broadcast ephemeris clock error time series as training samples according to the training sample duration; Normalizing the broadcast ephemeris orbit error time series and the broadcast ephemeris clock error time series; Construct a long short-term memory neural network model; The predicted arc length is set by the long short-term memory neural network model, and the broadcast ephemeris orbit error time series and the broadcast ephemeris clock error time series are trained and predicted; as well as The prediction results are denormalized to generate broadcast ephemeris orbit error prediction values ​​and broadcast ephemeris clock error prediction values.

4. The method for predicting the accuracy parameters of navigation satellite time-varying messages according to claim 3, characterized in that:The broadcast ephemeris orbit error prediction value and broadcast ephemeris clock error prediction value are obtained with a broadcast ephemeris update period of 1 hour and a sampling interval of 5 minutes. 12 sets of orbit radial direction error prediction values ​​R, orbit normal direction error prediction values ​​N, orbit tangential direction error prediction values ​​and broadcast ephemeris clock error prediction values ​​CLK are obtained every hour. The satellite orbit and clock error prediction error prediction values ​​Δ i It is expressed as the following formula: Here, i represents hours, i=1,2,…,24.

5. The method for predicting the accuracy parameters of navigation satellite time-varying messages according to claim 4, characterized in that: The broadcast ephemeris accuracy parameter level value includes the broadcast ephemeris orbit accuracy parameter level value SISAI oe , broadcast ephemeris clock error accuracy constant parameter level value SISAI ocb , broadcast ephemeris clock error accuracy velocity item parameter level value SISAI oc1 And the broadcast ephemeris clock error accuracy acceleration parameter level value SISAI oc2 , where the broadcast ephemeris orbit accuracy parameter level value SISAI oe , broadcast ephemeris clock error accuracy constant parameter level value SISAI ocb Take the upper limit value in the index relationship table, the broadcast ephemeris clock error accuracy speed item parameter level value SISAI oc1 And the broadcast ephemeris clock error accuracy acceleration parameter level value SISAI oc2 The transformation is expressed as follows: SIDE oc1 =-log2SISA oc1 -14 SIDE oc2 =-log2SISA oc1 -28.

6. A navigation satellite time-varying message accuracy parameter prediction system, characterized in that: include: A broadcast ephemeris error calculation module, which is configured to calculate the broadcast ephemeris orbit error and the broadcast ephemeris clock error according to the broadcast ephemeris and post-precision ephemeris of the satellite navigation system; A broadcast ephemeris error prediction module, configured to generate a broadcast ephemeris orbit error prediction value and a broadcast ephemeris clock error prediction value through a long short-term memory neural network algorithm according to the broadcast ephemeris orbit error and the broadcast ephemeris clock error; A broadcast ephemeris accuracy parameter calculation module, configured to calculate a broadcast ephemeris orbit accuracy parameter and a broadcast ephemeris clock error accuracy parameter according to the broadcast ephemeris orbit error prediction value and the broadcast ephemeris clock error prediction value; as well as a broadcast ephemeris accuracy parameter level value conversion module, configured to convert the broadcast ephemeris orbit accuracy parameter and the broadcast ephemeris clock error accuracy parameter into a broadcast ephemeris accuracy parameter level value; The broadcast ephemeris orbit error prediction value and the broadcast ephemeris clock error prediction value CLK of each epoch are obtained according to the sampling interval, wherein the broadcast ephemeris orbit error prediction value includes an orbit radial direction error prediction value R, an orbit normal direction error prediction value N, and an orbit tangential direction error prediction value T; The orbit accuracy parameters for calculating broadcast ephemeris include: The broadcast ephemeris orbit accuracy parameters are calculated based on the orbit tangent direction error prediction value T and the orbit normal direction error prediction value N. One broadcast ephemeris orbit accuracy parameter is calculated every hour, where the orbit tangent plane prediction accuracy SISA is calculated for each epoch within one hour. oe,i,j , and take the orbit tangent plane prediction accuracy within one hour SISA oe,i,j The maximum value is taken as the broadcast ephemeris orbit accuracy parameter SISA for that hour oe,i , expressed as the following formula: Where i represents the hour, j represents the sampling points within an hour, j = 1, 2, ..., 12; The parameters for calculating the broadcast ephemeris clock error accuracy include: Calculate the broadcast ephemeris clock error accuracy parameter RCLK for each error sampling point based on the orbit radial direction error prediction value R and the broadcast ephemeris clock error prediction value CLK i,j , expressed as the following formula: RCLK i,j =|R i,j |+|CLK i,j | Where i represents the hour, j represents the sampling points within an hour, j = 1, 2, ..., 12; Calculate the broadcast ephemeris clock error accuracy parameters of multiple sets of broadcast ephemeris, where the ground uploads one day of broadcast ephemeris to the satellite at a time, the update cycle of the broadcast ephemeris is one hour, and a total of 24 sets of broadcast ephemeris are uploaded; When calculating the broadcast ephemeris clock error accuracy parameters of each set of broadcast ephemeris, the clock error prediction error time series within one hour starting from the reference time t0 of the set of broadcast ephemeris is selected for fitting; If the reference time t0 is less than 26 hours from the current time, a first-order linear model is used to fit the constant term parameter SISA′ of the broadcast ephemeris clock error accuracy parameter. ocb,i and speed parameter SISA oc1,i If the reference time t0 is greater than or equal to 26 hours from the current time, a second-order linear model is used to fit the constant term parameter SISA′ of the broadcast ephemeris clock error accuracy parameter ocb,i , speed parameter SISA oc1,i And the acceleration parameter SISA oc2,i , expressed as the following formula: RCLK i,j =SISA' ocb,i +SISA oc1,i *(t j -t0) t-t 0,i ≤26h RCLK i,j =SISA' ocb,i +SISA oc1,i *(t-t0)+SISA oc2,i *(t j -t0) 2 t-t 0,i >26h Where i represents the i-th set of broadcast ephemeris, t represents the current calculation time, and t 0,i represents the reference time of the i-th set of broadcast ephemeris, t j represents the time of each clock error sampling point in the clock error prediction error time series used in calculating the i-th set of broadcast ephemeris; Calculate the mean square error information rms (RLCK) of the clock error prediction time series used in the i-th group of broadcast ephemeris i , expressed as the following formula: as well as Select the maximum value of the bias term and the mean square error information of the clock error time series in the fitting model as the bias term SISA of the i-th group of broadcast ephemeris clock error accuracy parameters ocb,i , expressed as the following formula: REMAINDER ocb,i =max[REMAINING' ocb,i ,rms(RCLK) i ]。