A method and system for taming a rubidium atomic clock based on a long-wave timing signal

Abstract

The present invention belongs to, but is not limited to, the field of time and frequency technology. It discloses a method and system for taming a rubidium atomic clock based on a long-wave timing signal. The method establishes a mathematical model for the long-wave timing signal, analyzes the time-frequency characteristics of the long-wave timing signal, uniformly analyzes and models the long-wave timing signal, which is affected by multiple factors, and corrects the long-wave time-varying signal, which exhibits significant fluctuations and periodic effects, into a stable timing signal. A time difference model for taming a rubidium atomic clock is constructed by superimposing a quadratic polynomial with periodic fluctuation terms, and corrects the original time difference data between the rubidium clock and the long-wave. The rubidium atomic clock is tamed, and the corrected time difference data between the rubidium clock and the long-wave is used in conjunction with the least squares method to achieve taming of the rubidium atomic clock. The present invention solves the problem of large short-term jitter and significant long-term periodic effects in the long-wave timing signal; reduces the volatility of the time difference data between the long-wave and rubidium atomic clocks; and expands the application of long-wave timing in taming atomic clocks.

Description

Technical Field

[0001] The present invention belongs to the field of time and frequency technology, and in particular relates to a method and system for taming a rubidium atomic clock based on a long-wave timing signal. Background Art

[0002] With the development of atomic frequency standard technology, the application of large-scale time and frequency systems in various fields, both domestically and internationally, has been limited in recent years by their high cost and demanding operating environments. Rubidium atomic clocks, with their competitive price, have become the preferred choice. However, rubidium atomic clocks suffer from frequency drift and aging issues, failing to meet the accuracy requirements of these systems. Therefore, it is necessary to tame rubidium atomic clock technology so that it can maintain high accuracy over a long period of time.

[0003] The GNSS (Global Navigation Satellite System) system provides accurate, 24 / 7 timing services at a low cost and with simple operation. Therefore, GNSS systems are often used in engineering projects to train rubidium atomic clocks. However, GNSS systems are susceptible to spoofing and have poor anti-interference capabilities. Furthermore, they cannot function effectively indoors, underground, or underwater due to limited signal transmission. To mitigate the risks of GNSS applications, technologies for training rubidium atomic clocks based on long-wave timing signals have gained attention in recent years. This patented long-wave timing system utilizes the Loran-C navigation system (Loran-C) timing architecture.

[0004] The long-wave timing system has become the best choice for supplementing and backing up satellite navigation systems due to its extremely strong anti-interference performance and relatively stable signal characteristics. Therefore, in order to meet the challenge of maintaining time and frequency accuracy in complex scenarios, especially to solve the time synchronization problem in the case of GNSS system denial or failure, it is of great significance to study the long-wave-based atomic clock taming algorithm, which will further promote the engineering application of land-based long-wave timing systems.

[0005] However, compared to GNSS signals, the propagation delay of long-wave signals is affected by numerous factors, including complex variations in electrical properties, and exhibits significant time-varying characteristics. Consequently, when using long-wave timing signals to tame rubidium atomic clocks, the time difference data collected suffers from significant jitter and long-term, significant periodic effects. Currently, long-wave timing signals are rarely used to tame rubidium atomic clocks in engineering. Therefore, an algorithm for tame rubidium atomic clocks based on long-wave timing signals is urgently needed. Summary of the Invention

[0006] In response to the problems existing in the prior art, the present invention provides a method for taming a rubidium atomic clock based on a long-wave timing signal. Based on this method, a corresponding rubidium atomic clock control system is designed, realizing the function of taming a rubidium atomic clock with a long-wave timing signal.

[0007] The present invention is implemented as follows: a method for taming a rubidium atomic clock based on a long-wave timing signal, comprising:

[0008] S1. Establish a mathematical model for the long-wave timing signal. By analyzing the time-frequency characteristics of the long-wave timing signal, the long-wave timing signal affected by various factors is uniformly analyzed and modeled. The long-wave timing signal is then corrected, and the long-wave time-varying signal with obvious fluctuations and periodic effects is corrected into a stable timing signal.

[0009] S2, based on the mathematical model of the long-wave timing signal, improve the rubidium atomic clock time difference data model, construct a time difference model that is a quadratic polynomial superimposed with periodic fluctuation terms tamed for the rubidium atomic clock, and correct the original time difference data of the rubidium clock and the long wave;

[0010] S3, taming the rubidium atomic clock, uses the corrected rubidium clock and long-wave time difference data combined with the least squares method to achieve the taming of the rubidium atomic clock.

[0011] The method for taming a rubidium atomic clock based on a long-wave timing signal specifically includes:

[0012] Step 1: mathematical model of long-wave timing signal delay based on trigonometric functions;

[0013] This study analyzes the frequency domain characteristics of the 1PPS (Pulse Per Second) signal output by a long-wave timing receiver and the UTC (Universal Time Coordinated) signal generated by the National Time Service Center (NTSC). The time difference data (LoranC-UTC) is converted from the time domain to the frequency domain using the Fast Fourier Transform (FFT). The authors then establish a mathematical model for the long-wave timing signal and modify the time difference data to convert the long-wave time-varying signal, which exhibits significant fluctuations and periodic effects, into a stable timing signal.

[0014] Step 2: Improve the rubidium atomic clock time difference data model and perform time difference data correction;

[0015] According to the analysis results of step 1, the time difference data between the 1PPS signal output by the rubidium atomic clock and the 1PPS signal output by the long-wave timing receiver also has a significant periodic fluctuation component. Therefore, based on the mathematical model of the long-wave timing signal, the rubidium atomic clock time difference model is transformed into a time difference model with a quadratic polynomial superimposed on a periodic fluctuation term. The original time difference data is used to solve the periodic term parameters in the time difference model, and then the data fitted by the periodic term is deducted from the original data to obtain the stable time difference data.

[0016] Step 3: Taming the rubidium atomic clock;

[0017] According to the processing results of step 2, the parameters of the rubidium atomic clock are estimated using the least squares method using the corrected stationary time difference data, and the performance parameters of the rubidium atomic clock are calculated. Then, the ping-pong control algorithm is used to adjust the adjustment amount of the rubidium atomic clock and the output frequency of the rubidium atomic clock, ultimately achieving the purpose of taming the rubidium atomic clock with the long-wave timing signal.

[0018] Furthermore, in the step 1, a multi-channel time interval counter is used to collect the 1PPS signal output by the UTC (NTSC) and the 1PPS signal output by the long-wave timing receiver.

[0019] Furthermore, the conversion formula for converting the time domain time difference data to the frequency domain using FFT in step 1 is:

[0020] ,

[0021] In the formula Indicates time difference data, Represents the frequency spectrum of the time difference data.

[0022] Furthermore, in step 1, a mathematical model of long-wave timing signal delay is established, which is expressed as follows:

[0023] ;

[0024] In the formula represents the number of periodic terms, represents the sampling time, represents the long wave delay data, represents the initial deviation, 、 、 Respectively represent The amplitude, frequency and phase of each cycle, represents the remaining residual.

[0025] Furthermore, in step 1, the time difference data of the long-wave timing signal is corrected:

[0026] ,

[0027] in, Indicates the corrected long wave delay data.

[0028] Furthermore, in step 2, the Rubidium-LoranC time difference model is transformed into a time difference model with a quadratic polynomial superimposed on a periodic fluctuation term, which is expressed as follows:

[0029] ,

[0030] in ， express The time difference between the 1PPS signal output by the rubidium atomic clock and the 1PPS signal output by the long-wave timing receiver at the moment, a represents the phase deviation at the initial moment, b represents the frequency deviation at the initial moment, and c represents the frequency drift. represents the remaining residual.

[0031] Furthermore, according to the FFT analysis results, if the time difference data contains multiple periodic terms, multiple trigonometric functions are constructed to transform the model The formula is simplified to matrix form:

[0032] ,

[0033] in represents the observed value, Represents the time difference data at the mth moment; represents the coefficient matrix, represents the mth moment, represents the frequency of the nth periodic term; represents the parameters to be estimated, represents the phase of the nth periodic term; represents the coefficient of the first periodic term; represents the coefficient of the nth periodic term;

[0034] Substitute the original time difference data into the matrix equation system and solve the matrix equation system to obtain the various parameters to be estimated, as shown in the following formula:

[0035] ;

[0036] From the original data By deducting the periodic fitting data from the equation, we can obtain the stable time difference data between the long wave and rubidium atomic clocks, as shown in the following formula:

[0037] ;

[0038] In the formula represents the corrected Rubidium-LoranC time difference data, Representation matrix The transposed matrix of represents the phase deviation at the initial moment; b represents the frequency deviation at the initial moment; c represents the frequency drift; represents the remaining residual, 、 、 Respectively represent The amplitude, frequency and phase of each cycle, express The time difference between the 1PPS signal output by the rubidium atomic clock and the 1PPS signal output by the long-wave timing receiver at the moment.

[0039] Furthermore, in step 3, the Rubidium-LoranC time difference data corrected in step 2 is used to tame the rubidium atomic clock, and a performance parameter estimation model of the rubidium atomic clock is established using the least squares method. It can be expressed in matrix form as:

[0040] ,

[0041] in represents the coefficient matrix, represents the mth sampling moment; Indicates the performance parameters of the rubidium clock to be determined; represents the phase deviation at the initial moment; b represents the frequency deviation at the initial moment; c represents the frequency drift; represents the remaining residual, 、 、 Respectively represent The amplitude, frequency and phase of each cycle, express The time difference between the 1PPS signal output by the rubidium atomic clock and the 1PPS signal output by the long-wave timing receiver;

[0042] By solving the matrix equations, the performance parameters of the rubidium atomic clock are obtained, and then the ping-pong control algorithm is used to adjust the adjustment amount of the rubidium atomic clock, ultimately realizing the long-wave timing signal taming the rubidium atomic clock frequency standard.

[0043] Another object of the present invention is to provide a method for training a rubidium atomic clock based on a long-wave timing signal and a system for training a rubidium atomic clock based on a long-wave timing signal, comprising a long-wave timing receiver, a rubidium atomic clock, a multi-channel time interval counter and a computer;

[0044] The output ends of the NTSC clock room, the long-wave timing receiver and the rubidium atomic clock are respectively connected to the input end of the multi-channel time interval counter, the output end of the multi-channel time interval counter is connected to the input end of the computer, and the output end of the computer is connected to the input end of the rubidium atomic clock.

[0045] Furthermore, the 1PPS signal output by the rubidium atomic clock, long-wave timing receiver and NTSC clock room is transmitted to a multi-channel time interval counter, and the time difference data output by the multi-channel time interval counter is transmitted to a computer, which adjusts the rubidium atomic clock based on the frequency adjustment amount.

[0046] First, the present invention establishes a mathematical model of long-wave timing signals. By constructing a time difference model consisting of a quadratic polynomial superimposed with periodic fluctuation terms, the volatility problem of the time difference data between long-wave and rubidium atomic clocks is effectively reduced. The invention also expands the application of long-wave timing in atomic clock training, providing reliable protection for the timing system to maintain time and frequency accuracy in complex scenarios, especially for solving time synchronization problems in the event of GNSS denial or failure.

[0047] Second, this invention innovatively proposes an algorithm for taming rubidium atomic clocks based on long-wave timing signals, directly addressing the pain point of the existing rubidium atomic clocks' insufficient long-term stability. By constructing a mathematical model of the long-wave timing signal, the algorithm deeply analyzes and accurately corrects its time-frequency characteristics, effectively smoothing out fluctuations and periodic effects caused by various factors, thereby significantly enhancing the signal's long-term stability. We employ a multi-channel time interval counter to synchronously capture the 1PPS signals of the UTC (NTSC) and long-wave timing receivers, and apply FFT technology to penetrate the frequency domain to accurately identify and calibrate the main periodic terms, further consolidating the foundation of the long-wave signal's stability.

[0048] On this basis, we optimized the rubidium atomic clock time difference data model, innovatively incorporating quadratic polynomials and periodic fluctuation terms to deeply purify the raw time difference data. By constructing and solving a system of matrix equations and accurately estimating the parameters using the least squares method, we successfully removed the interference of the periodic terms and achieved stabilization of the long-wave and rubidium atomic clock time difference data. This not only effectively addressed the frequency drift and periodic fluctuation issues of the rubidium atomic clock, but also significantly improved its time accuracy.

[0049] We then used the corrected Rubidium-LoranC time difference data and the least-squares method to construct a rubidium atomic clock performance parameter estimation model. This, supplemented by a ping-pong control algorithm for fine-tuning, achieved the ultimate calibration of the rubidium atomic clock frequency standard. This groundbreaking algorithm not only pushes the frequency accuracy of rubidium atomic clocks to a new high, but also comprehensively improves their short-term and long-term frequency stability, making them a shining example in the field of precision time and frequency measurement.

[0050] The core of this invention is the introduction of a long-wave timing signal calibration mechanism. This innovation fundamentally makes up for the shortcomings of existing rubidium atomic clocks in terms of time and frequency stability, and achieves a technological leap. Specifically, the 10MHz frequency accuracy is improved from -5E-11 in the free state to -2.27E-13, and the frequency stability reaches 7.39E-13 in 100 seconds, and is as low as 2.18E-13 at 10,000 seconds. These remarkable improvements not only greatly enhance the reliability and accuracy of rubidium atomic clocks, but also open up a broad space for them in higher-standard application scenarios, demonstrating the significant contribution and far-reaching impact of this invention in the field of time and frequency measurement. BRIEF DESCRIPTION OF THE DRAWINGS

[0051] Figure 1 This is a flow chart of a method for taming a rubidium atomic clock based on a long-wave timing signal provided by an embodiment of the present invention;

[0052] Figure 2 This is a block diagram of long-wave time difference data acquisition provided by an embodiment of the present invention;

[0053] Figure 3 This is an FFT analysis diagram of the LoranC-UTC time difference data provided by an embodiment of the present invention;

[0054] Figure 4 This is a comparison diagram before and after calibration of the long-wave timing signal provided by an embodiment of the present invention;

[0055] Figure 5 This is a comparison diagram of the long-wave timing signal stability before and after calibration provided by an embodiment of the present invention;

[0056] Figure 6 This is a block diagram of a rubidium atomic clock control system provided by an embodiment of the present invention;

[0057] Figure 7 This is a comparison chart of Rubidium-LoranC time difference data before and after processing provided by an embodiment of the present invention;

[0058] Figure 8 This is a diagram showing the frequency stability of a rubidium atomic clock after training, as provided in an embodiment of the present invention. DETAILED DESCRIPTION

[0059] In order to make the purpose, technical solutions and advantages of the present invention more clearly understood, the present invention is further described in detail below in conjunction with the embodiments. It should be understood that the specific embodiments described herein are only used to explain the present invention and are not intended to limit the present invention.

[0060] like Figure 1 As shown, the method for taming a rubidium atomic clock based on a long-wave timing signal provided by an embodiment of the present invention includes:

[0061] S1. Establish a mathematical model for the long-wave timing signal. By analyzing the time-frequency characteristics of the long-wave timing signal, the long-wave timing signal affected by various factors is uniformly analyzed and modeled. The long-wave timing signal is then corrected, and the long-wave time-varying signal with obvious fluctuations and periodic effects is corrected into a stable timing signal.

[0062] S2, based on the mathematical model of the long-wave timing signal, improve the rubidium atomic clock time difference data model, construct a time difference model that is a quadratic polynomial superimposed with periodic fluctuation terms tamed for the rubidium atomic clock, and correct the original time difference data of the rubidium clock and the long wave;

[0063] S3, taming the rubidium atomic clock, uses the corrected rubidium clock and long-wave time difference data combined with the least squares method to achieve the taming of the rubidium atomic clock.

[0064] The following are specific steps of a method for taming a rubidium atomic clock based on a long-wave timing signal provided by an embodiment of the present invention.

[0065] 1. Establishing a mathematical model for long-wave timing signals

[0066] 1.1 Building a time difference data collection platform

[0067] The platform uses a multi-channel time interval counter to collect the 1PPS signal output by UTC (NTSC) and the 1PPS signal output by the long wave timing receiver. Figure 2 Make the connections as shown.

[0068] 1.2 Analysis of long-wave timing signal characteristics

[0069] To better demonstrate and analyze the characteristics of longwave data, we analyze the frequency domain characteristics of the 1PPS signal output by the longwave timing receiver and the 1PPS signal time difference data (LoranC-UTC) output by UTC (NTSC). We use FFT to convert the time domain time difference data to the frequency domain to observe whether it has obvious periodic characteristics. The conversion formula is:

[0070] (1);

[0071] In the formula Indicates time difference data, Represents the frequency spectrum of the time difference data.

[0072] Pick Figure 2 Taking the LoranC-UTC time difference data of one day as an example, the frequency domain characteristics are analyzed as follows Figure 3 As shown, it can be seen that there are obvious periodic items in the LoranC-UTC time difference data, the most obvious of which is the 24-hour period, corresponding to Figure 3For point A in the figure, if the vertical axis amplitude of 10 is used as the threshold, two other periodic terms of 12 hours and 8 hours can be obtained, corresponding to points B and C in the figure respectively.

[0073] 1.3 Establishing a mathematical model for long-wave timing signals

[0074] Based on the above analysis, a mathematical model of long-wave timing signal delay based on trigonometric functions is established, which is expressed as follows:

[0075] (2);

[0076] In the formula represents the number of periodic terms, represents the sampling time, represents the long wave delay data, represents the initial deviation, 、 、 Respectively represent The amplitude, frequency and phase of each cycle, represents the remaining residual.

[0077] 1.4 Correction of time difference data of long-wave timing signal

[0078] Based on the analysis in 1.3, the time difference data of the long-wave timing signal can be corrected using the following formula:

[0079] (3);

[0080] in, Indicates the corrected long wave delay data.

[0081] by Figure 3 The vertical axis amplitude 10 is the threshold value, that is, , the comparison results before and after removing the three periodic terms are as follows Figure 4 As shown in Table 1, it can be seen that the stability of the long-wave timing signal is effectively improved by establishing a trigonometric function model, especially the long-term stability is improved from the previous 2.97E-12 (20000s) to 5.16E-13 (20000s), and from 2.34E-12 (40000s) to 2.37E-13 (40000s), as shown in Table 1 and Figure 5 As shown, Figure 5 The blue curve represents the stability after calibration, and the purple curve represents the stability before calibration.

[0082] Table 1 Comparison of stability before and after long-wave calibration

[0083] Allan variance Stability before long wave calibration Stability after long wave calibration 1s 1.21E-8 1.21E-8 100s 2.89E-11 2.89E-11 1000s 5.72E-12 5.71E-12 10000s 2.73E-12 1.21E-12 20000s 2.97E-12 5.16E-13 40000s 2.34E-12 2.37E-13

[0084] 2. Improving the Rubidium Atomic Clock Time Difference Data Model and Correcting Time Difference Data

[0085] According to the results of step one, on the basis of the mathematical model of long-wave timing signal, an algorithm for taming the rubidium atomic clock with long-wave timing signal is proposed.

[0086] 2.1. Establishing a long-wave and rubidium clock time difference data model

[0087] To achieve the goal of taming the rubidium atomic clock with a long-wave timing signal, we must first establish a time difference data model between the long-wave and rubidium atomic clocks. Usually, the time difference between the rubidium atomic clock and Coordinated Universal Time (UTC) exhibits the following pattern:

[0088] (4);

[0089] The time difference model parameters include: Represents the rubidium clock error data, the constant term a represents the phase deviation at the initial moment, the first-order term parameter b represents the frequency deviation at the initial moment, and the second-order term parameter c represents the frequency drift and represents random noise.

[0090] according to Figure 6 Connect as shown to collect the original time difference data of the 1PPS signal output by the rubidium atomic clock and the 1PPS signal output by the long-wave timing receiver.

[0091] 2.2. Establishing an improved model for long-wave and rubidium clock time difference data

[0092] Based on the analysis of the long-wave timing signal characteristics in step 1, it can be inferred that the time difference data (Rubidium-LoranC) between the 1PPS signal output by the rubidium atomic clock and the 1PPS signal output by the long-wave timing receiver contains both the characteristics of the rubidium atomic clock and the long-wave characteristics. Therefore, the Rubidium-LoranC time difference model can be transformed into a time difference model consisting of a quadratic polynomial superimposed on a periodic fluctuation term, expressed as follows:

[0093] (5);

[0094] in express The time difference between the 1PPS signal output by the rubidium atomic clock and the 1PPS signal output by the long-wave timing receiver at the moment, represents the phase deviation at the initial moment, b represents the frequency deviation at the initial moment, and c represents the frequency drift. represents the remaining residual.

[0095] According to the FFT analysis results, if the time difference data contains multiple periodic terms, multiple trigonometric functions are constructed to simplify the model into a matrix form:

[0096] (6)

[0097] in represents the observed value; Represents the time difference data at the mth moment; represents the coefficient matrix; represents the mth moment; represents the frequency of the nth periodic term; represents the parameter to be estimated; represents the phase of the nth periodic term; represents the coefficient of the first periodic term; represents the coefficient of the nth periodic term;

[0098] Substitute the original time difference data into the matrix equations and solve the matrix equations, as shown in Equation (7), to obtain the parameters to be estimated:

[0099] (7);

[0100] From the original data By deducting the periodic fitting data from the equation, we can obtain the stable long-wave and rubidium atomic clock time difference data, as shown in equation (7):

[0101] (8);

[0102] in The modified Rubidium-LoranC time difference data is shown in the figure below. Figure 7 shown.

[0103] Taming the Rubidium Atomic Clock

[0104] The Rubidium-LoranC time difference data corrected in step 2 is used to tame the rubidium atomic clock, and the performance parameter estimation model of the rubidium atomic clock is established using the least squares method. (8) can be expressed in matrix form as:

[0105] (9);

[0106] in represents the coefficient matrix, represents the mth sampling moment; represents the desired rubidium clock performance parameters, a represents the phase deviation at the initial moment; b represents the frequency deviation at the initial moment; c represents the frequency drift.

[0107] By solving the matrix equations, the performance parameters of the rubidium atomic clock are obtained. Then, the ping-pong control algorithm is used to adjust the adjustment amount of the rubidium atomic clock, and finally the long-wave timing signal is used to tame the rubidium atomic clock frequency standard. The frequency accuracy of the rubidium atomic clock outputting 10MHz is improved from -5E-11 (free state) to -2.27E-13; the frequency stability is 7.39E-13 (100s), 2.18E-13 (10000s), as shown in Figure 2. Figure 8 The stability comparison with the free-state rubidium atomic clock is shown in Table 2.

[0108] Table 2 Comparison of stability of rubidium atomic clock before and after taming

[0109] Allan variance Stability after taming Free Rubidium Atomic Clock (Instruction Manual) 1s 1.69E-11 2E-11 10s 4.10E-12 1E-11 100s 7.39E-13 2E-12 1000s 2.76E-13 10000s 2.18E-13 20000s 1.63E-13

[0110] Application Example 1: Satellite Navigation System

[0111] In satellite navigation systems, high-precision time synchronization is a key factor in ensuring positioning accuracy. Traditional rubidium atomic clocks suffer from poor long-term stability due to frequency drift and periodic fluctuations, which impacts the overall performance of the system. The method of the present invention precisely calibrates the long-wave timing signal with the rubidium atomic clock, significantly improving the long-term stability and frequency accuracy of the rubidium atomic clock. With this technology, the rubidium atomic clock in the satellite navigation system can provide a more accurate time reference, reduce the accumulation of time errors, and improve the positioning accuracy and reliability of the entire navigation system. Furthermore, this improvement can reduce satellite maintenance costs and extend the service life of the rubidium atomic clock.

[0112] Application Example 2: Communication Network Synchronization

[0113] In modern communication networks, high-precision time synchronization is crucial for data transmission, network coordination, and system performance. In particular, in 5G networks and future 6G networks, time synchronization accuracy at the microsecond or even nanosecond level becomes a necessary condition. The method of the present invention significantly improves the time and frequency stability of the rubidium atomic clock by introducing a long-wave timing signal to tame the rubidium atomic clock. After being applied to communication network synchronization, it can ensure high-precision time synchronization between network nodes, improve the stability and accuracy of data transmission, reduce delays and jitter, and meet the needs of high-bandwidth, low-latency communication. This is of great significance for fields such as financial transactions, smart grids, and the Internet of Things that have extremely high requirements for time synchronization.

[0114] It should be noted that the embodiments of the present invention can be implemented by hardware, software, or a combination of software and hardware. The hardware portion can be implemented using dedicated logic; the software portion can be stored in a memory and executed by an appropriate instruction execution system, such as a microprocessor or dedicated design hardware. Those skilled in the art will understand that the above-mentioned devices and methods can be implemented using computer-executable instructions and / or contained in processor control code, for example, such as a carrier medium such as a disk, CD or DVD-ROM, a programmable memory such as a read-only memory (firmware), or a data carrier such as an optical or electronic signal carrier. The device and its modules of the present invention can be implemented by hardware circuits such as very large-scale integrated circuits or gate arrays, semiconductors such as logic chips, transistors, etc., or programmable hardware devices such as field programmable gate arrays, programmable logic devices, etc., can also be implemented by software executed by various types of processors, or can be implemented by a combination of the above-mentioned hardware circuits and software, such as firmware.

[0115] The above description is only a specific embodiment of the present invention, but the scope of protection of the present invention is not limited thereto. Any modifications, equivalent substitutions and improvements made by any technician familiar with this technical field within the technical scope disclosed by the present invention and within the spirit and principles of the present invention should be covered by the scope of protection of the present invention.

Claims

1. A method for taming a rubidium atomic clock based on a long-wave timing signal, characterized in that: include: Step 1: A mathematical model of the long-wave timing signal delay based on trigonometric functions is established. The frequency domain characteristics of the 1PPS signal output by the long-wave timing receiver and the UTC generated by NTSC and the 1PPS signal time difference data LoranC-UTC output by UTC are analyzed. Fast Fourier transform (FFT) is used to convert the time domain time difference data to the frequency domain to observe whether the 1PPS signal output by the long-wave timing receiver has obvious periodic characteristics. A mathematical model of the long-wave timing signal is established, and the time difference data of the long-wave timing signal is corrected to correct the long-wave time-varying signal with obvious fluctuation and periodic effects into a stable timing signal. Step 2: Improve the rubidium atomic clock time difference data model and correct the time difference data; according to the analysis results of step 1: the time difference data of the 1PPS signal output by the rubidium atomic clock and the 1PPS signal output by the long-wave timing receiver also have obvious periodic fluctuation components. Based on the mathematical model of the long-wave timing signal, the rubidium atomic clock time difference model is transformed into a time difference model with a quadratic polynomial superimposed on a periodic fluctuation term; use the original time difference data to solve the periodic term parameters in the time difference model, and then deduct the periodic term fitting data from the original data to obtain stable time difference data; Step three, tame the rubidium atomic clock; according to the processing results of step two, use the corrected stationary time difference data to use the least squares method to estimate the parameters of the rubidium atomic clock, calculate the performance parameters of the rubidium atomic clock, and then use the ping-pong control algorithm to adjust the adjustment amount of the rubidium atomic clock and adjust the output frequency of the rubidium atomic clock, ultimately achieving the purpose of taming the rubidium atomic clock with the long-wave timing signal.

2. The method for training a rubidium atomic clock based on a long-wave timing signal according to claim 1, wherein: In the step 1, a multi-channel time interval counter is used to collect the 1PPS signal output by the UTC (NTSC) and the 1PPS signal output by the long-wave timing receiver.

3. The method for training a rubidium atomic clock based on a long-wave timing signal according to claim 1, wherein: The conversion formula for converting the time domain time difference data to the frequency domain using FFT in step 1 is: ， In the formula Indicates time difference data, Represents the frequency spectrum of the time difference data.

4. The method for training a rubidium atomic clock based on a long-wave timing signal according to claim 1, wherein: The long-wave timing signal delay mathematical model based on trigonometric functions established in step 1 is expressed as follows: ； In the formula represents the number of periodic terms, represents the sampling time, represents the long wave delay data, represents the initial deviation, 、 、 Respectively represent The amplitude, frequency and phase of each cycle, represents the remaining residual.

5. The method for training a rubidium atomic clock based on a long-wave timing signal according to claim 1, wherein: In the step 1, the time difference data of the long-wave timing signal is corrected using the following formula: ， in, represents the corrected long-wave delay data, represents the number of periodic terms, Indicates the sampling time, the value is t=1,2,3......, represents the long wave delay data, 、 、 Respectively represent The amplitude, frequency and phase of each cycle.

6. The method for training a rubidium atomic clock based on a long-wave timing signal according to claim 1, wherein: In step 2, the Rubidium-LoranC time difference model is transformed into a time difference model with a quadratic polynomial superimposed on a periodic fluctuation term, which is expressed as follows: ， in ， express The time difference between the 1PPS signal output by the rubidium atomic clock and the 1PPS signal output by the long-wave timing receiver at the moment, represents the phase deviation at the initial moment, b represents the frequency deviation at the initial moment, and c represents the frequency drift. represents the remaining residual, 、 、 Respectively represent The amplitude, frequency and phase of each cycle, represents the number of periodic terms, Indicates the sampling time.

7. The method for training a rubidium atomic clock based on a long-wave timing signal according to claim 6, wherein: According to the FFT analysis results, if the time difference data contains multiple periodic terms, construct multiple trigonometric functions and transform the model The formula is simplified to matrix form: ， in represents the observed value, Represents the time difference data at the mth moment; represents the coefficient matrix, represents the mth moment, represents the frequency of the nth periodic term; represents the parameter to be estimated; represents the phase of the nth periodic term; represents the coefficient of the first periodic term; represents the coefficient of the nth periodic term; Substitute the original time difference data into the matrix equation system and solve the matrix equation system to obtain the various parameters to be estimated, as shown in the following formula: ； From the original data By deducting the periodic fitting data from the equation, we can obtain the stable time difference data between the long wave and rubidium atomic clocks, as shown in the following formula: ； In the formula represents the corrected Rubidium-LoranC time difference data, Representation matrix The transposed matrix of represents the phase deviation at the initial moment; b represents the frequency deviation at the initial moment; c represents the frequency drift; represents the remaining residual, 、 、 Respectively represent The amplitude, frequency and phase of each cycle, express The time difference between the 1PPS signal output by the rubidium atomic clock and the 1PPS signal output by the long-wave timing receiver at the moment.

8. The method for training a rubidium atomic clock based on a long-wave timing signal according to claim 1, wherein: In the step 3, the Rubidium-LoranC time difference data corrected in the step 2 is used to tame the rubidium atomic clock, and a performance parameter estimation model of the rubidium atomic clock is established using the least squares method. It can be expressed in matrix form as: ， in represents the coefficient matrix, represents the mth sampling moment; Indicates the performance parameters of the rubidium clock to be determined; represents the phase deviation at the initial moment; b represents the frequency deviation at the initial moment; c represents the frequency drift; represents the remaining residual, 、 、 Respectively represent The amplitude, frequency and phase of each cycle, express The time difference between the 1PPS signal output by the rubidium atomic clock and the 1PPS signal output by the long-wave timing receiver; By solving the matrix equations, the performance parameters of the rubidium atomic clock are obtained, and then the ping-pong control algorithm is used to adjust the adjustment amount of the rubidium atomic clock, ultimately realizing the long-wave timing signal taming the rubidium atomic clock frequency standard.

9. A system for training a rubidium atomic clock based on a long-wave timing signal, according to any one of claims 1 to 8, characterized in that: It includes a long-wave timing receiver, a rubidium atomic clock, a multi-channel time interval counter and a computer; The output ends of the NTSC clock room, the long-wave timing receiver and the rubidium atomic clock are respectively connected to the input end of the multi-channel time interval counter, the output end of the multi-channel time interval counter is connected to the input end of the computer, and the output end of the computer is connected to the input end of the rubidium atomic clock.

10. The system for training a rubidium atomic clock based on a long-wave timing signal according to claim 9, characterized in that: The 1PPS signal output by the rubidium atomic clock, long-wave timing receiver and NTSC clock room is transmitted to a multi-channel time interval counter, and the time difference data output by the multi-channel time interval counter is transmitted to a computer, which adjusts the rubidium atomic clock based on the frequency adjustment amount.

Images