A method and apparatus for calculating atomic time based on hydrogen atomic clock drift prediction

Abstract

This application discloses an atomic time calculation method and apparatus based on hydrogen atomic clock drift prediction. The method includes: obtaining historical clock difference data from a database; obtaining initial clock difference values ​​between the combined clock and a reference clock and between each clock and the combined clock based on the clock difference data of each atomic clock; performing time function fitting based on the clock difference data and the drift values ​​between each clock and the combined clock to obtain a time-varying frequency drift prediction value; and subtracting the frequency drift value introduced by the hydrogen atomic clock in the combined clock based on the time-varying frequency drift prediction value to obtain an atomic timescale. This application solves the problem of long-term frequency drift in atomic timescale calculation based on hydrogen atomic clocks, thereby enabling real-time updates of atomic clock frequency drift during atomic timescale calculation, allowing for accurate subtraction of drift terms, and ultimately obtaining an atomic timescale that simultaneously possesses short-term and long-term stability.

Description

Technical Field

[0001] This application relates to the field of atomic clocks, and more specifically, to a method and apparatus for calculating atomic time based on the drift prediction of a hydrogen atomic clock. Background Technology

[0002] Atomic time calculation is a time-keeping method used in timekeeping systems. A timekeeping system typically consists of multiple atomic clocks, each generating an atomic time scale. Since each atomic clock can malfunction, collecting clock difference data from each clock and using statistical methods to calculate the combined atomic time scale plays a crucial role in establishing the atomic time scale of the timekeeping system, ensuring its high stability.

[0003] Atomic clocks in timekeeping systems generally include hydrogen atomic clocks and cesium atomic clocks. The difference between cesium and hydrogen atomic clocks lies in the short-term stability of hydrogen atomic clocks. However, due to significant frequency drift during long-term operation, hydrogen atomic clocks have poorer long-term stability than cesium atomic clocks. Currently, commonly used atomic time algorithms include ALGOS, AT1, and the Kalman algorithm. The Kalman algorithm is based on the Kalman filter and does not use the classic weight calculation method. The ALGOS algorithm only considers the influence of frequency difference changes in the atomic clock during atomic time calculation, making it more suitable for atomic time calculations based on cesium atomic clocks. The AT1 algorithm considers the frequency drift of hydrogen atomic clocks in atomic time calculations, but treats it as a constant derived from historical data during the calculation process, without considering that the frequency drift term of hydrogen atomic clocks changes over time. This results in the calculated time scale not accurately subtracting the frequency drift term, and the final result still contains frequency drift. Summary of the Invention

[0004] This application provides an atomic time calculation method and apparatus based on hydrogen atomic clock drift prediction, to at least solve the problem of long-term frequency drift in atomic timescale calculation based on hydrogen atomic clock.

[0005] According to one aspect of this application, an atomic time calculation method based on hydrogen atomic clock drift prediction is provided, comprising: obtaining historical clock difference data from a database, wherein the clock difference data includes date and time data and clock difference data between each atomic clock and a reference clock; obtaining initial clock difference values ​​between each clock and the reference clock and between each clock and a combined clock based on the clock difference data of each atomic clock; performing time function fitting based on the clock difference data and the clock difference values ​​between each clock and the combined clock to obtain a frequency drift prediction value that varies with time; and obtaining the clock difference between the combined clock and the reference clock after deducting the atomic clock speed and clock drift based on the clock difference data, the frequency difference, and the frequency drift prediction value that varies with time, thereby obtaining an atomic time scale.

[0006] Furthermore, based on the clock difference data of each atomic clock, the initial clock difference values ​​between each clock and the combined clock are obtained. The initial clock difference values ​​between the combined clock and the reference clock are obtained by averaging and weighting the clock difference data of each atomic clock.

[0007] Furthermore, based on the clock difference data and the clock difference values ​​between the combined clock and the reference clock, a time function is fitted to obtain the predicted frequency drift value that varies with time: the clock difference at the next moment is estimated based on the clock difference, frequency difference, and frequency drift at the previous moment; the weight value corresponding to each atomic clock is calculated based on the clock difference at the next moment for each atomic clock; the clock difference data between the combined clock and the reference clock is calculated based on the weight value; and the frequency drift prediction value that varies with time is obtained by fitting a time function to the drift values ​​between each clock and the reference clock.

[0008] Furthermore, subtracting the frequency drift value introduced by the hydrogen atomic clock of the combined clock based on the predicted frequency drift value that varies with time includes: obtaining the predicted frequency drift value of the hydrogen atomic clock at the next moment based on the predicted frequency drift value that varies with time; and subtracting the predicted frequency drift value of the hydrogen atomic clock at the next moment from the hydrogen atomic clock of the combined clock.

[0009] Further, obtaining the predicted frequency drift value of the hydrogen atomic clock at the next moment based on the predicted frequency drift value that changes over time includes: using the first difference of the frequency difference for exponential filtering to calculate the calculated frequency drift value; calculating the predicted frequency drift value based on the predicted frequency drift value that changes over time; and adding the calculated frequency drift value and the predicted frequency drift value with equal weights to obtain the predicted frequency drift value of the hydrogen atomic clock at the next moment.

[0010] According to another aspect of this application, an atomic time calculation device based on hydrogen atomic clock drift prediction is also provided, comprising: an acquisition module for acquiring historical clock difference data from a database, wherein the clock difference data includes date and time data and clock difference data between each atomic clock and a reference clock; a obtaining module for obtaining initial clock difference values ​​between each clock and the combined clock and the combined clock and the reference clock based on the clock difference data of each atomic clock; a fitting module for performing time function fitting based on the clock difference data and the clock difference values ​​between each clock and the combined clock to obtain a predicted frequency drift value that varies with time; and a subtraction module for subtracting the frequency drift value introduced by the hydrogen atomic clock of the combined clock based on the predicted frequency drift value that varies with time to obtain an atomic timescale.

[0011] Furthermore, the obtaining module is used to: obtain the initial clock difference value between each clock and the combined clock based on the clock difference data; and to obtain the initial clock difference value between the combined clock and the reference clock by averaging and weighting the clock difference data of each atomic clock.

[0012] Furthermore, the fitting module is used to: estimate the clock difference between each clock and the combined clock at the next moment based on the clock difference, frequency difference, and frequency drift at the previous moment; calculate the weight value corresponding to each atomic clock based on the clock difference at the next moment of each atomic clock; and perform time function fitting on the drift value between each clock and the combined clock based on the clock difference at the next moment of each atomic clock to obtain the predicted value of frequency drift that changes with time.

[0013] Furthermore, the subtraction module is used to: obtain the predicted frequency drift value of the hydrogen atomic clock at the next moment based on the predicted frequency drift value that changes over time; and subtract the predicted frequency drift value of the hydrogen atomic clock at the next moment from the hydrogen atomic clock of the combined clock.

[0014] Furthermore, the subtraction module is used for: using the first difference of the frequency difference for exponential filtering to obtain the frequency drift calculation value; calculating the frequency drift prediction value based on the frequency drift prediction value that changes with time; and adding the frequency drift calculation value and the frequency drift prediction value with equal weight to obtain the frequency drift prediction value of the hydrogen atomic clock at the next moment.

[0015] In this embodiment, historical clock difference data is obtained from a database. This clock difference data includes date and time data and clock difference data between each atomic clock and a reference clock. Initial clock difference values ​​between each atomic clock and the combined clock, as well as between the combined clock and the reference clock, are obtained based on the clock difference data of each atomic clock. A time function is fitted using the clock difference data and the clock difference values ​​between each atomic clock and the combined clock to obtain a predicted frequency drift value that varies over time. The frequency drift value introduced by the hydrogen atomic clock in the combined clock is subtracted based on the predicted frequency drift value that varies over time to obtain the atomic timescale. This application solves the problem of long-term frequency drift in atomic timescale calculation based on hydrogen atomic clocks, enabling real-time updates of atomic clock frequency drift during atomic timescale calculation, allowing for accurate subtraction of drift terms, and ultimately obtaining an atomic timescale that possesses both short-term and long-term stability. Attached Figure Description

[0016] The accompanying drawings, which form part of this application, are used to provide a further understanding of this application. The illustrative embodiments and descriptions of this application are used to explain this application and do not constitute an undue limitation of this application. In the drawings:

[0017] Figure 1 This is a flowchart of atomic time calculation based on hydrogen atomic clock drift prediction according to an embodiment of this application;

[0018] Figure 2 This is a flowchart of an atomic time calculation method based on hydrogen atomic clock drift prediction according to an embodiment of this application. Detailed Implementation

[0019] It should be noted that, unless otherwise specified, the embodiments and features described in this application can be combined with each other. This application will now be described in detail with reference to the accompanying drawings and embodiments.

[0020] It should be noted that the steps shown in the flowchart in the accompanying drawings can be executed in a computer system such as a set of computer-executable instructions, and although a logical order is shown in the flowchart, in some cases the steps shown or described may be executed in a different order than that shown here.

[0021] This embodiment provides a method for calculating atomic time based on hydrogen atomic clock drift prediction. Figure 1 This is a flowchart of atomic time calculation based on hydrogen atomic clock drift prediction according to an embodiment of this application. The following is a description of... Figure 1 The process involves explaining the steps involved.

[0022] Step S102: Obtain historical clock difference data from the database, wherein the clock difference data includes date and time data and clock difference data between each atomic clock and the reference clock;

[0023] Step S104: Obtain the initial clock difference values ​​between each atomic clock and the combined clock, as well as the initial clock difference values ​​between the combined clock and the reference clock, based on the clock difference data of each atomic clock. For example, the initial clock difference values ​​between the combined clock and the reference clock can be obtained by averaging and weighting the clock difference data of each atomic clock.

[0024] Step S106: Based on the clock difference data and the clock difference values ​​between individual clocks and the combined clock, a time function is fitted to obtain a predicted value of frequency drift that varies with time. For example, the clock difference at the next moment can be estimated based on the clock difference, frequency difference, and frequency drift at the previous moment; the weight value corresponding to each atomic clock is calculated based on the clock difference at the next moment of each atomic clock; and a time function is fitted to the drift values ​​between individual clocks and the combined clock based on the clock difference at the next moment of each atomic clock to obtain a predicted value of frequency drift that varies with time.

[0025] Step S108: Subtract the frequency drift value introduced by the hydrogen atomic clock of the combined clock from the predicted frequency drift value that changes with time to obtain the atomic time scale.

[0026] In this step, the predicted frequency drift value of the hydrogen atomic clock at the next moment can be obtained based on the predicted frequency drift value that varies with time; the predicted frequency drift value of the hydrogen atomic clock at the next moment is subtracted from the hydrogen atomic clock of the combined clock. Optionally, obtaining the predicted frequency drift value of the hydrogen atomic clock at the next moment based on the predicted frequency drift value that varies with time includes: using the first difference of the frequency difference for exponential filtering to calculate the calculated frequency drift value; calculating the predicted frequency drift value based on the predicted frequency drift value that varies with time; and adding the calculated frequency drift value and the predicted frequency drift value with equal weights to obtain the predicted frequency drift value of the hydrogen atomic clock at the next moment.

[0027] The above steps solve the problem of long-term frequency drift in atomic timescale calculation based on hydrogen atomic clocks. This allows for real-time updates of the frequency drift of the atomic clock during atomic timescale calculation, enabling accurate subtraction of the drift term and ultimately obtaining an atomic timescale that possesses both short-term and long-term stability.

[0028] The following description uses an optional embodiment. This optional embodiment provides an atomic time calculation method based on hydrogen atomic clock drift prediction. This method records the clock difference data of each hydrogen atomic clock in real time, updates the combined clock difference value over time, and simultaneously updates the frequency difference value and clock drift value. The following description uses... Figure 2 The steps in this method are explained.

[0029] S1: Extract clock difference data collected by the timekeeping laboratory from the database, including date and time data and clock difference data between each atomic clock and the reference clock.

[0030] S2: Perform data preprocessing on the extracted data, including outlier removal and data averaging.

[0031] S3: Obtain the initial clock difference between the combined clock and the reference clock, as well as the initial clock difference between each individual clock and the reference clock, using the average weighting method; obtain the initial frequency difference and initial frequency drift term respectively using the historical data fitting method; set the initial weights of each clock to be equal.

[0032] S4: Based on the AT1 algorithm, an improved atomic timescale calculation method is adopted, and an exponential filtering method based on time function constraints is used to obtain the accurate prediction value of frequency drift introduced by the hydrogen atomic clock.

[0033] S5: During the atomic timescale calculation, the frequency drift value introduced by the hydrogen atomic clock is updated in real time, and the atomic timescale after accurate frequency drift subtraction is finally obtained.

[0034] In this optional embodiment, the specific implementation process includes estimating the clock difference for the next moment based on the clock difference, frequency difference, and clock drift of the previous moment; calculating the weight value of each clock for the next moment based on the expected error of each clock; and updating the frequency difference value for the next moment using an exponential filtering method. Specifically, an exponential filtering method based on time function constraints is used to update the hydrogen clock frequency drift value for the next moment. Specifically: the first difference of the frequency difference is used for exponential filtering to obtain the calculated frequency drift value; the predicted frequency drift value is obtained by calculating the time function fitted based on historical frequency difference data. The calculated frequency drift value and the predicted frequency drift value are added together with equal weights to obtain the accurate predicted clock drift value for the next moment.

[0035] The first-order difference formula for the frequency difference is:

[0036]

[0037] The exponential filtering formula for calculating the frequency drift value is as follows:

[0038]

[0039] The formula for calculating the accurate frequency difference prediction value is as follows:

[0040]

[0041] Among them, y i (t) represents the clock speed of the i-th clock at time t. and Let represent the calculated and predicted frequency drift values ​​of the i-th clock at time (t+τ), respectively. The corresponding time interval τ is obtained by minimizing the Allan variance. minUsed for exponential frequency averaging time constant m i The calculation.

[0042] The m i The calculation formula is:

[0043]

[0044] The atomic time calculation method based on hydrogen atomic clock drift prediction in this optional embodiment can realize real-time update of the frequency drift of the atomic clock during the atomic time scale calculation process, so that the drift term is accurately deducted, and finally an atomic time scale with both short-term stability and long-term stability is obtained.

[0045] The atomic time calculation method based on hydrogen atomic clock drift prediction provided in this optional embodiment may include raw clock difference data extraction, clock difference data preprocessing, and algorithm calculation to update the atomic time scale. Specifically, the atomic time calculation method described in this example improves upon the existing AT1 algorithm by employing an exponential filtering method based on a time function to calculate the frequency drift of the hydrogen atomic clock in real time, thereby accurately subtracting the frequency drift error introduced by the hydrogen atomic clock from the obtained atomic time scale and achieving higher long-term stability.

[0046] First, clock difference data collected by the timekeeping laboratory was extracted from the database, including date and time data and clock difference data between each atomic clock and the reference clock.

[0047] Then, the extracted data is preprocessed, including outlier removal and data averaging.

[0048] The initial clock difference between the combined clock and the reference clock was obtained by means of an average weighting method; the initial frequency difference and the initial frequency drift term were obtained by means of historical data fitting method; and the initial weights of each clock were set to be equal.

[0049] Assume the clock difference between the i-th clock and the combined clock is x. i The frequency difference is y i Zhong Piao is z i The clock difference between the i-th clock and the reference clock is x. ij The clock difference value of the next time step (t+1 time step) is predicted based on the clock difference, frequency difference, and frequency drift of the previous time step (t time step), as shown in formula (1).

[0050]

[0051] Furthermore, the clock difference x between the combined clock and the reference clock is obtained. r And the calculated value x of the clock difference between each individual clock and the combination clock. i See formulas (2) and (3).

[0052]

[0053] x i (t+τ)=x ij (t+τ)-x r (t+τ) (3)

[0054] Among them, w i (t+τ) represents the weight value of each atomic clock at time (t+τ). In the AT1 algorithm, the weight update method relies on the error between the predicted and calculated clock difference values ​​of each clock and the combined clock. Under the maximum authority condition, the weight value of each clock is inversely correlated with its corresponding expected error. Specifically:

[0055]

[0056]

[0057]

[0058]

[0059] The frequency difference at time (t+τ) is calculated using an exponential filtering method. The clock difference data at time (t+τ) is obtained by first difference and then exponential filtering, as shown in formulas (8)-(9).

[0060]

[0061]

[0062] Where, m i The time constant is the average time constant of the exponential frequency. The corresponding time interval τ is obtained by minimizing the Allan variance. min Used for exponential frequency averaging time constant m i The calculation.

[0063] The m i The calculation formula is:

[0064]

[0065] Specifically, the frequency drift value at time (t+τ) is calculated using exponential filtering. This is achieved by first-order difference of the frequency difference data at time (t+τ) followed by exponential filtering, as shown in formulas (11)-(12). The predicted frequency drift value at time (t+τ) is obtained by least-squares fitting of frequency difference data over a historical period.

[0066]

[0067]

[0068] By combining the calculated and predicted values ​​of the clock drift with equal weights, the final predicted value of the clock drift at time (t+τ) can be obtained.

[0069]

[0070] Among them, y i (t) represents the clock speed of the i-th clock at time t. and Let represent the calculated and predicted frequency drift values ​​of the i-th clock at time (t+τ), respectively.

[0071] This embodiment provides an atomic time calculation method based on hydrogen atomic clock drift prediction, which ensures that the frequency drift term introduced by the hydrogen atomic clock is updated in real time during the atomic time calculation process, thereby enabling the calculated atomic time to accurately deduct the frequency drift error.

[0072] In this embodiment, an electronic device is provided, including a memory and a processor. The memory stores a computer program, and the processor is configured to run the computer program to perform the methods described in the above embodiments.

[0073] The aforementioned program can run on a processor or be stored in memory (or computer-readable medium). Computer-readable medium includes both permanent and non-permanent, removable and non-removable media, and information storage can be achieved by any method or technology. Information can be computer-readable instructions, data structures, program modules, or other data. Examples of computer storage media include, but are not limited to, phase-change memory (PRAM), static random access memory (SRAM), dynamic random access memory (DRAM), other types of random access memory (RAM), read-only memory (ROM), electrically erasable programmable read-only memory (EEPROM), flash memory or other memory technologies, CD-ROM, digital versatile optical disc (DVD) or other optical storage, magnetic tape, magnetic magnetic disk storage or other magnetic storage devices, or any other non-transferable medium that can be used to store information accessible by a computing device. As defined herein, computer-readable medium does not include transient computer-readable media, such as modulated data signals and carrier waves.

[0074] These computer programs may also be loaded onto a computer or other programmable data processing equipment to cause a series of operational steps to be performed on the computer or other programmable equipment to produce a computer-implemented process, thereby providing instructions that execute on the computer or other programmable equipment for implementing the process. Figure 1 One or more processes and / or boxes Figure 1The steps for the functions specified in one or more boxes can be implemented using different modules, and different steps can be implemented using different modules.

[0075] This embodiment provides such an apparatus or system. The apparatus, referred to as an atomic time calculation device based on hydrogen atomic clock drift prediction, includes: an acquisition module for acquiring historical clock difference data from a database, wherein the clock difference data includes date and time data and clock difference data between each atomic clock and a reference clock; a obtaining module for obtaining a frequency drift value between the combined clock and the reference clock based on the clock difference data of each atomic clock; a fitting module for performing time function fitting based on the clock difference data and the drift value between the combined clock and the reference clock to obtain a predicted frequency drift value that varies with time; and a subtraction module for subtracting the frequency drift value introduced by the hydrogen atomic clock of the combined clock based on the predicted frequency drift value that varies with time, to obtain an atomic timescale.

[0076] The system or apparatus is used to implement the functions of the methods in the above embodiments. Each module in the system or apparatus corresponds to each step in the method, as has been described in the method and will not be repeated here.

[0077] For example, the obtaining module is used to: average and weight the clock difference data of each atomic clock to obtain the initial clock difference value between the combined clock and the reference clock. Optionally, the fitting module is used to: estimate the clock difference at the next moment based on the clock difference, frequency difference, and clock drift of the previous moment; calculate the weight value corresponding to each atomic clock based on the clock difference at the next moment of each atomic clock; and perform time function fitting on the drift value between each clock and the combined clock based on the clock difference at the next moment of each atomic clock to obtain the predicted value of frequency drift that changes with time.

[0078] For example, the subtraction module is used to: obtain the predicted frequency drift value of the hydrogen atomic clock at the next moment based on the predicted frequency drift value that varies with time; and subtract the predicted frequency drift value of the hydrogen atomic clock at the next moment from the combined clock's hydrogen atomic clock. Optionally, the subtraction module is used to: use the first difference of the frequency difference for exponential filtering to calculate the calculated frequency drift value, and calculate the predicted frequency drift value based on the predicted frequency drift value that varies with time; and add the calculated frequency drift value and the predicted frequency drift value with equal weights to obtain the predicted frequency drift value of the hydrogen atomic clock at the next moment.

[0079] The atomic time calculation method based on hydrogen atomic clock drift prediction provided by the above embodiments is simple and easy to implement, and can make the final atomic timescale achieve high short-term and long-term stability. It is only necessary to add the frequency drift calculation value obtained by exponential filtering and the frequency drift prediction value obtained by fitting the time function based on historical frequency difference data with equal weight to obtain the clock drift prediction value that changes with time.

[0080] The above are merely embodiments of this application and are not intended to limit the scope of this application. Various modifications and variations can be made to this application by those skilled in the art. Any modifications, equivalent substitutions, improvements, etc., made within the spirit and principles of this application should be included within the scope of the claims of this application.

Claims

1. A method for atomic time calculation based on hydrogen atomic clock drift prediction, characterized in that, The method comprises the following steps: obtaining historical clock difference data from a database, wherein the clock difference data comprises date and time data and clock difference data between each atomic clock and a reference clock; obtaining initial clock difference values between a combined clock and the reference clock and initial clock difference values between each atomic clock and the combined clock according to the clock difference data of each atomic clock; estimating the clock difference at the next moment according to the clock difference, frequency difference and frequency difference drift at the previous moment; calculating the weight value corresponding to each atomic clock according to the clock difference of each atomic clock at the next moment; calculating the clock difference data between the combined clock and the reference clock according to the weight value; and performing time function fitting on the drift value between each atomic clock and the reference clock to obtain a frequency difference drift prediction value varying with time; subtracting the frequency difference drift value introduced by the hydrogen atomic clock of the combined clock according to the frequency difference drift prediction value varying with time to obtain an atomic time scale; the step of subtracting the frequency difference drift value introduced by the hydrogen atomic clock of the combined clock according to the frequency difference drift prediction value varying with time comprises: obtaining a frequency difference drift prediction value of the hydrogen atomic clock at the next moment according to the frequency difference drift prediction value varying with time; and subtracting the frequency difference drift prediction value of the hydrogen atomic clock at the next moment from the hydrogen atomic clock of the combined clock.

2. The method of claim 1, wherein, the step of obtaining initial clock difference values between the combined clock and the reference clock and initial clock difference values between each atomic clock and the combined clock according to the clock difference data of each atomic clock comprises: averaging and weighting the clock difference data of each atomic clock to obtain the initial clock difference values between the combined clock and the reference clock.

3. The method of claim 1, wherein, the step of obtaining a frequency difference drift prediction value of the hydrogen atomic clock at the next moment according to the frequency difference drift prediction value varying with time comprises: performing one-order difference of the frequency difference to obtain a frequency drift calculation value by exponential filtering calculation, and obtaining a frequency drift prediction value according to the frequency difference drift prediction value varying with time; and adding the frequency drift calculation value and the frequency drift prediction value equally to obtain the frequency difference drift prediction value of the hydrogen atomic clock at the next moment.

4. An atomic time calculation device based on hydrogen atomic clock drift prediction, characterized by, The method comprises the following steps: an obtaining module, configured to obtain historical clock difference data from a database, wherein the clock difference data comprises date and time data and clock difference data between each atomic clock and a reference clock; an obtaining module, configured to obtain initial clock difference values between each atomic clock and a combined clock and initial clock difference values between the combined clock and the reference clock according to the clock difference data of each atomic clock; a fitting module, configured to estimate the clock difference between each atomic clock and the combined clock at the next moment according to the clock difference at the previous moment, frequency difference and frequency drift; calculate the weight value corresponding to each atomic clock according to the clock difference of each atomic clock at the next moment; and perform time function fitting on the drift value between each atomic clock and the combined clock according to the clock difference of each atomic clock at the next moment to obtain a frequency difference drift prediction value varying with time; a subtracting module, configured to subtract the frequency difference drift value introduced by the hydrogen atomic clock of the combined clock according to the frequency difference drift prediction value varying with time to obtain an atomic time scale; and the step of subtracting the frequency difference drift value introduced by the hydrogen atomic clock of the combined clock according to the frequency difference drift prediction value varying with time comprises: obtaining a frequency difference drift prediction value of the hydrogen atomic clock at the next moment according to the frequency difference drift prediction value varying with time; and subtracting the frequency difference drift prediction value of the hydrogen atomic clock at the next moment from the hydrogen atomic clock of the combined clock.

5. The apparatus of claim 4, wherein, the obtaining module is configured to: The initial value of the clock difference between each clock and the combined clock is obtained according to the clock difference data; and the initial value of the clock difference between the combined clock and the reference clock is obtained by averaging and weighting the clock difference data of each atomic clock.

6. The apparatus of claim 4, wherein, The deduction module is used for: The first difference of the frequency difference is used for exponential filtering calculation to obtain a frequency drift calculation value, and a frequency drift prediction value is calculated according to the frequency difference drift prediction value changing with time; The frequency drift calculation value and the frequency drift prediction value are added equally to obtain the frequency difference drift prediction value of the hydrogen atomic clock at the next moment.

Images