A short wave time signal propagation delay prediction method and device

Abstract

The application particularly relates to a short-wave time signal propagation time delay prediction method and device, wherein the method is based on prior data to train a short-wave time signal propagation time delay prediction model, does not need ionosphere data as support, and only needs several simple multiplication operations and one logarithm operation after a user obtains model parameters, does not involve solution of differential equations, and is simple to realize. In addition, since the prediction model fits time variation of the ionosphere, not only can high time delay prediction precision be maintained, but also the parameter model is unchanged within the same day time, only needs to be obtained once, and after real-time measurement of signal level, the signal propagation time delay can be calculated, and the parameter model does not need to be frequently obtained.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of short wave time service, in particular to a short wave time service signal propagation time delay prediction method and device, and a computer program product. BACKGROUND

[0002] In the related art, short wave time service is a time service means using radio frequencies in the short wave band as a carrier to broadcast standard time and standard frequency, which has the characteristics of long distance, simple receiving equipment and strong anti-destruction. BPM short wave time service system is constructed and operated by the National Time Service Center of the Chinese Academy of Sciences, and is the earliest major scientific and technological infrastructure in China. Since 1970, the system has been broadcasting national standard time and standard frequency at 2.5MHz, 5MHz, 10MHz and 15MHz four frequency points alternately.

[0003] In order to realize long-distance short wave time service, the short wave time service signal (sky wave) transmitted from the sending end must pass through the continuous refraction-reflection of the ionosphere before reaching the long-distance receiving end. The ionosphere is not only an important part of the earth's atmosphere, but also the most complex structure and the most closely related atmospheric layer to short wave sky wave transmission. The instability of sky wave propagation caused by the complex changes of the ionosphere is one of the main factors affecting the accuracy of short wave time service.

[0004] The existing method for calculating the propagation time delay of short wave time service signal with high accuracy is the ray tracing method, which usually needs to obtain sufficient ionospheric data (such as ionospheric layered structure, electron density, critical frequency, etc.) from the International Reference Ionosphere Model (IRI Model), and also involves solving a large number of differential equations in the ray trajectory calculation process. Therefore, for short wave timing terminal users who require simple receiving equipment, there are problems such as difficulty in obtaining ionospheric data and high computational complexity in calculating the propagation time delay of short wave time service signal using this method.

[0005] It should be noted that the information disclosed in the above background section is only used to strengthen the understanding of the background of the present application, and therefore can include information that does not constitute prior art known to those of ordinary skill in the art. SUMMARY

[0006] The present application provides a short wave time service signal propagation time delay prediction method and device, and a computer program product, which can effectively overcome the defects in the prior art.

[0007] Other characteristics and advantages of the present application will become apparent from the following detailed description, or will be learned by practice of the present application.

[0008] According to a first aspect of the present application, a short wave time service signal propagation time delay prediction method is provided, the method comprising:

[0009] collecting prior data of a target time length, performing time period division and preprocessing on the prior data according to a preset rule to obtain sample data corresponding to each time period, and constructing a sample data set; wherein the sample data of each time period includes a propagation time delay of the short-wave time service signal in the time period corresponding to the time period and a signal level;

[0010] updating parameters of a short-wave time service signal propagation time delay prediction model using the sample data set to determine updated model hyperparameters; wherein the short-wave time service signal propagation time delay prediction model is constructed based on a carrier frequency feature of the short-wave time service signal, a signal level feature of the short-wave time service signal at a current time, and a time period feature at the current time; and the model hyperparameters include hyperparameters corresponding to the signal level feature of the short-wave time service signal at the current time and the time period feature at the current time.

[0011] broadcasting the updated model hyperparameters for use in updating the short-wave time service signal propagation time delay prediction model by a short-wave time service receiving terminal according to the updated model hyperparameters, and using the updated model to predict the propagation time delay of the short-wave time service signal.

[0012] In some example embodiments, the method further comprises: constructing a short-wave time service signal propagation time delay prediction model in advance according to a carrier frequency feature of the short-wave time service signal, a signal level feature of the short-wave time service signal at a current time, and a time period feature corresponding to the current time, including:

[0013]

[0014] wherein, is a date serial number d , t is a signal propagation time delay prediction value at the time; is a date serial number d , t is a signal level of the short-wave time service signal at the time, used to represent a signal level feature of the short-wave time service signal at the current time; is t a time period corresponding to the time, used to represent a time period feature corresponding to the current time; represents a carrier frequency feature of the short-wave time service signal; , , are model hyperparameters, respectively.

[0015] In some example embodiments, the method further comprises: configuring the carrier frequency feature of the short-wave time service signal according to a carrier frequency of the short-wave time service signal , including:

[0016]

[0017] wherein, fA carrier frequency of the short-wave time signal.

[0018] In some example embodiments, prior data of a target time length is collected, the prior data is divided into time periods and preprocessed according to a preset rule to obtain sample data corresponding to each time period, including:

[0019] Signal propagation time delay measurement values of the short-wave time signal and signal level measurement values of the short-wave time signal corresponding to several dates before the current date are collected;

[0020] The data is sorted by date, and a corresponding date serial number is configured d

[0021] Each day is divided into M time periods, and a corresponding time period serial number λ is configured for each time period;

[0022] The arithmetic mean value of the signal propagation time delay measurement values in each time period is calculated, and is configured as the signal propagation time delay corresponding to the time period;

[0023] The arithmetic mean value of the signal level measurement values in each time period is calculated, and is configured as the signal level corresponding to the time period.

[0024] In some example embodiments, the method further includes: configuring the value range of the number of time periods M as: , and divisible by 86400; wherein, .

[0025] In some example embodiments, the short-wave time signal propagation time delay prediction model is parameter updated using the sample data set to determine the updated model hyperparameters, including:

[0026] The sample data set is divided into a training set and a validation set;

[0027] The model parameter vector is defined according to the model hyperparameters ; wherein, ;

[0028] The signal delay variable is defined according to the signal time delay of each time period, and the signal delay feature is constructed in combination with the signal delay variable and the carrier frequency feature ; and the signal delay feature vector is configured according to the signal delay feature ; wherein, , ; The carrier frequency feature of the short-wave time signal is represented;

[0029] The signal level variable is defined according to the signal level of each time period, and the signal level matrix is constructed in combination with the signal level variable and the time period feature ; wherein, ​

[0030]

[0031] wherein, denotes a time period, and M denotes a number of time periods; is t an average value of signal levels of short wave time signal in the time period;

[0032] a signal level matrix corresponding to the training set is utilized , a signal delay feature vector , and a pre-configured model order n, a starting date serial number of the training set to determine a model parameter vector , including: ;

[0033] The model parameter vector is substituted into the time delay prediction model to obtain a corresponding propagation time delay prediction value; and an error between the time delay prediction value and the time delay measurement value is calculated.

[0034] The model parameter vector is verified by using a verification set, and the model parameter vector corresponding to the minimum average absolute error is configured as the updated model hyperparameter.

[0035] In some example embodiments, configuring the model parameter vector corresponding to the minimum average absolute error as the updated model hyperparameter includes:

[0036]

[0037] wherein, denotes the model parameter vector adopted in calculating the time delay prediction value when the error is the minimum value; ; is a signal propagation time delay prediction value of the time period of the day with the date serial number ; is a signal propagation time delay of the time period of the day with the date serial number

[0038] .

[0039] In some example embodiments, the updated model is utilized to predict the propagation time delay of the short wave time signal, including:

[0040] The short wave signal level corresponding to the current short wave time signal and the time period corresponding to the current short wave time signal are input into the updated model to obtain the short wave time signal propagation time delay of the current short wave time signal output by the model.According to a second aspect of the present application, a short-wave time signal propagation delay prediction device is provided, comprising:

[0041] a time-reproducing device configured to provide a reference 10MHz signal and a reference 1PPS for a short-wave time monitoring receiver;

[0042] a short-wave time monitoring receiver configured to receive a short-wave time signal, and perform timing, time difference measurement, and signal level measurement on the received short-wave time signal, and output the time difference measurement data and the signal level measurement data to a host computer, wherein the time difference measurement data is used to convert into a corresponding propagation delay measurement value in the host computer;

[0043] a host computer configured to collect prior data of a target time length, divide and preprocess the prior data according to a preset rule to obtain sample data corresponding to each time period, and construct a sample data set, wherein the sample data of each time period includes a propagation delay measurement value and a signal level measurement value of the short-wave time signal in the time period, and the sample data set is used to update parameters of a short-wave time signal propagation delay prediction model to determine updated model hyperparameters, wherein the short-wave time signal propagation delay prediction model is constructed based on a carrier frequency feature of the short-wave time signal, a signal level feature of the short-wave time signal at a current time, and a time period feature at the current time, and the model hyperparameters include hyperparameters corresponding to the signal level feature of the short-wave time signal at the current time and the time period feature at the current time;

[0044] a parameter injection unit configured to broadcast the updated model hyperparameters, so that a short-wave time receiving terminal updates the short-wave time signal propagation delay prediction model according to the updated model hyperparameters, and uses the updated model to predict the propagation delay of the short-wave time signal.

[0045] According to a third aspect of the present application, a computer program product is provided, which stores a computer program, and the computer program is executed by a processor to implement the short-wave time signal propagation delay prediction method described above.

[0046] According to a fourth aspect of the present application, an electronic device is provided, comprising:

[0047] a processor; and

[0048] a memory configured to store executable instructions of the processor;

[0049] wherein the processor is configured to implement the short-wave time signal propagation delay prediction method described above by executing the executable instructions.

[0050] According to a fifth aspect of the present application, a storage medium is provided, which stores a computer program, and the computer program is executed by a processor to implement the short-wave time signal propagation delay prediction method described above.

[0051] The short-wave time signal propagation time delay prediction method provided by the embodiment of the present application utilizes the carrier frequency characteristics of the short-wave time signal, the signal level characteristics of the short-wave time signal at the current time, and the time period characteristics at the current time to construct a short-wave time signal propagation time delay prediction model, which linearly fits the logarithm of the signal propagation time delay and the signal level; and by using the signal level characteristics and the time period characteristics in the prediction model, nonlinear fitting of the ionospheric time variation is realized, which can maintain a high time delay prediction accuracy; in addition, by introducing the carrier frequency characteristics of the short-wave time signal in the prediction model, the signal attenuation problem of the time signal in the transmission process is fully considered, which further improves the accuracy of the time delay prediction at different times. In addition, the short-wave time signal propagation time delay prediction model is trained and the hyperparameters are updated by using the prior data in a period of time, which does not need ionospheric data support, and the parameter model is unchanged in the same day, so that the user only needs to obtain the parameter model once and measure the signal level in real time to calculate the signal propagation time delay, without the need to frequently obtain the parameter model.

[0052] It should be understood that the foregoing general description and the following detailed description are only exemplary and explanatory, and cannot limit the present application. BRIEF DESCRIPTION OF DRAWINGS

[0053] The drawings incorporated into the specification and forming a part thereof show embodiments consistent with the present application and, together with the specification, serve to explain the principles of the application. Obviously, the drawings in the following description are only some embodiments of the present application, and other drawings can be obtained by those skilled in the art without creative labor on the basis of these drawings.

[0054] Figure 1 The schematic diagram of a short-wave time signal propagation time delay prediction method according to an exemplary embodiment of the present application is shown schematically;

[0055] Figure 2 The schematic diagram of a time delay prediction method flow according to an exemplary embodiment of the present application is shown schematically;

[0056] Figure 3 The schematic diagram of a short-wave time signal propagation time delay prediction device according to an exemplary embodiment of the present application is shown schematically;

[0057] Figure 4 The schematic diagram of data processing of a propagation time delay prediction device according to an exemplary embodiment of the present application is shown schematically;

[0058] Figure 5 The schematic diagram of the principle of a data transmission unit according to an exemplary embodiment of the present application is shown schematically;

[0059] Figure 6 FIG. 1 schematically illustrates a configuration of an electronic device according to an example embodiment of the present disclosure. DETAILED DESCRIPTION

[0060] Example embodiments now will be described more fully hereinafter with reference to the accompanying drawings. Example embodiments, however, can be implemented in many different forms and should not be construed as limited to the examples set forth herein; rather, these embodiments are provided so that this disclosure will be thorough and complete, and will fully convey the scope of the example embodiments to those skilled in the art. Features, structures, or characteristics described in connection with one embodiment can be combined in any suitable manner in one or more embodiments.

[0061] In addition, the drawings are to be considered in all respects as illustrative and not restrictive; identical reference numerals have been used, where possible, to denote identical or similar parts that serve the same or similar purposes, and a repeated description thereof will be omitted. Some of the block diagrams shown in the drawings are functional entities that do not necessarily have to correspond to physically or logically independent entities. These functional entities can be implemented in the form of software, or in one or more hardware modules or integrated circuits, or in different network and / or processor devices and / or microcontroller devices.

[0062] In view of the disadvantages and deficiencies of the prior art, a short-wave time signal propagation delay prediction method is provided in the example embodiments, which is applied to a short-wave time system. Referring to Figure 1 As shown, the method can specifically include:

[0063] At step S11, prior data of a target time length is collected, the prior data is divided into time periods and preprocessed according to a preset rule to obtain sample data corresponding to each time period, and a sample data set is constructed; wherein the sample data of each time period includes a propagation delay of a short-wave time signal in the time period and a signal level;

[0064] At step S12, the sample data set is used to update parameters of a short-wave time signal propagation delay prediction model to determine updated model hyperparameters; wherein the short-wave time signal propagation delay prediction model is constructed based on a carrier frequency feature of the short-wave time signal, a signal level feature of the short-wave time signal at a current time, and a time period feature at the current time; and the model hyperparameters include hyperparameters corresponding to the signal level feature of the short-wave time signal at the current time and the time period feature at the current time.

[0065] At step S13, the updated model hyperparameters are broadcasted, so as to be used for a short-wave time receiving terminal to update the short-wave time signal propagation delay prediction model according to the updated model hyperparameters, and to use the updated model to predict a propagation delay of a short-wave time signal.

[0066] The method solves the problems of difficulty in obtaining ionospheric data and high complexity in calculation in the prior art applied to a simple short-wave timing terminal user of a receiving device.

[0067] Next, the steps of the short-wave time signal propagation time delay prediction method in the example embodiment will be described in more detail in combination with the accompanying drawings and examples.

[0068] In the example embodiment, the time delay prediction method can include: constructing a short-wave time signal propagation time delay prediction model in advance according to a carrier frequency feature of the short-wave time signal, a signal level feature of the short-wave time signal at a current time, and a time period feature corresponding to the current time, including:

[0069]

[0070] wherein, is a date sequence d, t is a signal propagation time delay prediction value at the time, in units of μs; is a date sequence d, t is a signal level of the short-wave time signal at the time, used to represent the signal level feature of the short-wave time signal at the current time; λ(t) is t is a time period corresponding to the time, used to represent the time period feature corresponding to the current time; b f represents the carrier frequency feature of the short-wave time signal; a 0, a 1, a i+1 are model hyperparameters, and ; n represents the model order.

[0071] wherein, the carrier frequency feature of the short-wave time signal is configured according to the carrier frequency of the short-wave time signal b f , including: b f =145-lg f ; wherein, f represents the carrier frequency of the short-wave time signal, in units of MHz.

[0072] Specifically, the time delay prediction method described above can be executed by an intelligent terminal of a short-wave time system. A propagation time delay prediction model applied to the short-wave time signal can be constructed in advance in the intelligent terminal. The model linearly fits the logarithm of the signal propagation time delay with the signal level, and nonlinearly fits the ionospheric time variation. The carrier frequency feature b fThe signal level amplification can eliminate the problem that the model parameters need more decimal points for representation and consume more bits for text arrangement. Meanwhile, the reduction of the division operation can reduce the calculation complexity of the terminal user.

[0073] Alternatively, in some example embodiments, b f = 145-lg f Since the model parameters already contain the constant term related to the signal level, the constant term in the formula does not affect the training of the model.

[0074] The use of the signal level feature and the time period feature corresponding to the current time in the prediction model realizes the nonlinear fitting of the ionospheric time variation. The introduction of the carrier frequency feature of the short-wave time signal in the prediction model fully considers the attenuation of the time signal in the transmission process,

[0075] In step S11, the prior data of the target time length is collected, the prior data is divided into time periods and preprocessed according to a preset rule to obtain sample data corresponding to each time period, and a sample data set is constructed. The sample data of each time period includes the propagation delay of the short-wave time signal in the time period and the signal level.

[0076] In an example, step S11 can include:

[0077] Step S21, collecting the signal propagation delay measurement value of the short-wave time signal and the signal level measurement value of the short-wave time signal corresponding to several dates before the current date;

[0078] Step S22, sorting the data by date and configuring the corresponding date serial number d;

[0079] Step S23, dividing each day into M time periods and configuring the corresponding time period serial number ;

[0080] Step S24, calculating the arithmetic mean value of the signal propagation delay measurement value in each time period and configuring it as the signal propagation delay corresponding to the time period;

[0081] Step S25, calculating the arithmetic mean value of the signal level measurement value in each time period and configuring it as the signal level corresponding to the time period.

[0082] Specifically, in an update cycle of a model hyperparameter, signal propagation delay measurement values (unit: μs) and signal level measurement values (unit: dBμV) of short wave time service signals of several days before the current date can be collected first. For example, the prior data of the previous three days can be obtained after the short wave time service signals are received and measured in real time by the short wave time service monitoring stations deployed far away from the short wave time service broadcasting system.

[0083] Specifically, the prior data can be sorted according to the acquisition date from small to large, and the corresponding date serial number is denoted by d; the signal propagation delay measurement values of each day are uniformly divided into M time periods with equal intervals starting from UTC time 0:00:00, and each time period is assigned a time period serial number. The arithmetic mean of the signal propagation delay measurement values in each time period is calculated, and the calculation result is taken as the signal propagation delay of the time period, thereby obtaining a signal propagation delay data table.

[0084] The signal level measurement values of each day are uniformly divided into M time periods with equal intervals starting from UTC time 0:00:00, and each time period is assigned a time period serial number, and then the arithmetic mean of the signal level measurement values in each time period is calculated, and the calculation result is taken as the signal level of the time period, thereby obtaining a signal level data table.

[0085] Wherein, the row number of the signal propagation delay data table and the signal level data table is the date serial number, and the column number is the time period serial number.

[0086] The signal propagation delay of the λth time period in the signal propagation delay data table with the date serial number d can be expressed as:

[0087]

[0088] Wherein, represents the signal propagation delay measurement value of the tth moment with the date serial number d.

[0089] The signal level of the λth time period in the signal level data table with the date serial number d can be expressed as:

[0090]

[0091] Wherein, represents the signal level measurement value of the tth moment with the date serial number d; the initial value of t is UTC time 0:00:00, and the value range is [0, 86399] unit: second.

[0092] In step S12, the parameters of the shortwave time signal propagation delay prediction model are updated using the sample dataset to determine the updated model hyperparameters. The shortwave time signal propagation delay prediction model is constructed based on the carrier frequency characteristics of the shortwave time signal, the signal level characteristics of the shortwave time signal at the current time, and the time period characteristics at the current time. The model hyperparameters include the hyperparameters corresponding to the signal level characteristics of the shortwave time signal at the current time and the time period characteristics at the current time.

[0093] For example, step S12 described above may include:

[0094] Step S31: Divide the sample dataset into a training set and a validation set;

[0095] Step S32: Define the model parameter vector based on the model hyperparameters. A ;in, ;

[0096] Step S33: Define delay variables based on signal delay for each time period, and construct signal delay features y by combining delay variables and carrier frequency characteristics. λ And configure the signal delay feature vector according to the signal delay characteristics. Y ;in, , ;

[0097] Step S34: Define signal level variables based on the signal levels of each time period, and construct a signal level matrix V by combining the signal level variables and time period characteristics;

[0098] Step S35: Utilize the signal level matrix V and signal delay feature vector corresponding to the training set. Y And the pre-configured model order n, the start date number of the training set, and so on. d s Determine the model parameter vector A, including: ;

[0099] Step S36, convert the model parameter vector A Substitute the values ​​into the delay prediction model to obtain the corresponding propagation delay prediction values; calculate the error between the delay prediction values ​​and the delay measurement values.

[0100] Step S37, use the validation set to test the model parameter vector. A To verify this, the model parameter vector corresponding to the minimum mean absolute error is configured as the updated model hyperparameters.

[0101] Specifically, signal propagation delay and signal level are selected from the propagation delay data table and the signal level data table, respectively, to construct the first... The sample set for the time period is denoted as ; and the validation set, denoted as , and is expressed as follows:

[0102]

[0103]

[0104] wherein, is the date serial number of the current date, is the date serial number of the day before the current date.

[0105] is expressed as The time period serial number corresponding to the time of day can be expressed as:

[0106]

[0107] Specifically, when the model hyperparameters are updated, the training set of the first time period in the sample set is selected, denoted as , and is expressed as follows:

[0108]

[0109] wherein, is the starting date serial number of the training set.

[0110] The time delay variable is defined and expressed as:

[0111]

[0112] wherein, is the signal time delay corresponding to each time period; is the geometric mean value calculated based on

[0113] The signal level variable is defined and expressed as:

[0114]

[0115] wherein, is the signal level corresponding to each time period; is the arithmetic mean value calculated based on The model parameter vector composed of the model hyperparameters is defined as:

[0116] . The signal delay feature vector

[0117] is defined and expressed as ; wherein the signal delay feature is constructed according to the time delay variable and the carrier frequency feature, including: . ​​​

[0118] Combination signal level variable , time interval, define signal level matrix , expressed as:

[0119]

[0120] Wherein, Indicates the time interval, M represents the maximum time interval value; t The average value of the signal level of the short wave time signal in the time interval.

[0121] After the given order n and the starting date sequence number , the corresponding matrix And the matrix Can be calculated by the above formula.

[0122] According to the And the , the model parameter vector can be calculated, and the formula is expressed as:

[0123]

[0124] Then, the elements in the currently calculated model parameter vector Is substituted into the prediction model, that is, when the model order is n and the training set starting date sequence number is , the signal propagation delay prediction value of the date sequence number , the first Time interval, the formula is expressed as:

[0125]

[0126] Adjust the values of the model order n And the starting date sequence number ; wherein, n The adjustment range of N , The adjustment range of , every time n Or Adjust once, repeat the above method to get the corresponding signal propagation delay prediction value.

[0127] Then, the verification set Can be used for verification, and the optimal model parameter is obtained by using the minimum average absolute error criterion, which is recorded as , expressed as:

[0128]

[0129] Wherein, Indicates when​ the signal propagation delay prediction value of the signal in the time period the model parameter vector used the signal propagation delay prediction value of the signal in the time period , the signal propagation delay prediction value of the signal in the time period the signal propagation delay prediction value of the signal in the time period , the signal propagation delay prediction value of the signal in the time period

[0130] In the training process, by adjusting n and ds (different n and ds correspond to different model parameters ), the average absolute error is minimized, and when it is minimized, the corresponding model parameter is the optimal model parameter .

[0131] The average absolute error belongs to the regression model evaluation index formula in the field of statistics and machine learning, which is used to quantify the average absolute error between the predicted value and the true value, and its advantages are as follows:

[0132] 1) Simple and intuitive calculation, by taking the absolute value of the deviation between the predicted value and the true value and then taking the arithmetic mean, the calculation process is not complex;

[0133] 2) Robust to outliers, since the absolute value is used, it does not amplify the influence of outliers like the root mean square error, and the result better reflects the true average error level of the data, and its use in the model training process can avoid the influence of large outliers on the model parameter training.

[0134] For example, in the process of fitting the ionospheric time variation with a polynomial, in order to avoid underfitting and overfitting as much as possible, and at the same time taking into account the calculation complexity, N=5 can be configured, that is, the upper limit value of the model order n is 5.

[0135] In step S13, the updated model hyperparameters are broadcasted for the short-wave time service receiving terminal to update the short-wave time service signal propagation delay prediction model according to the updated model hyperparameters, and to perform short-wave time service signal propagation delay prediction using the updated model.

[0136] For example, the updated model is used to perform short-wave time service signal propagation delay prediction, which includes:

[0137] The short-wave signal level corresponding to the current short-wave time service signal and the time period corresponding to the current short-wave time service signal are input into the updated model to obtain the short-wave time service signal propagation delay corresponding to the current short-wave time service signal output by the model.

[0138] ​​​Specifically, after determining the updated model hyperparameters for the current period, the time synchronization system can broadcast the model hyperparameters, enabling the shortwave time synchronization receiving terminal to receive the updated model hyperparameters for the current period, thereby realizing the local update of the time delay prediction model at the shortwave time synchronization receiving terminal.

[0139] After receiving a shortwave time signal, the shortwave time receiver terminal determines the corresponding signal level value. This indicates that the shortwave terminal user is on the current date. The signal level value measured at any given time.

[0140] Shortwave time synchronization terminal equipment, on the current date of the first day The signal level value is measured at each moment, and then the current signal level value is... The time period corresponding to the current moment Updated model hyperparameters Substituting this into the prediction model, we can calculate the current date's... The predicted signal propagation delay at time t is expressed as:

[0141]

[0142] in, , , for The elements in.

[0143] The method provided in the embodiments of the present invention is referred to Figure 2 As shown, the process includes: S201, acquiring several prior data, including propagation delay measurement data and signal level measurement data of shortwave time synchronization signals; S202, preprocessing the prior data to obtain a propagation delay data table and a signal level data table; S203, constructing a sample set and a validation set composed of signal propagation delay and signal level; S204, constructing a shortwave time synchronization signal propagation delay prediction model, which linearly fits the logarithm of signal propagation delay to the signal level and nonlinearly fits it to ionospheric time variation; S205, selecting a training set from the sample set to train the model parameters and obtain the optimal model parameters; S206, the terminal device calculates the signal propagation delay for each time period of the current date based on the signal level measurement values ​​for each time period of the current date and the optimal model parameters through the prediction model.

[0144] The method is based on prior data to train a short-wave time signal propagation time delay prediction model, does not need ionosphere data as support, after a user obtains model parameters, only needs several simple multiplication operations and a logarithm operation, does not involve solving of a differential equation, is simple to realize. In addition, since the prediction model fits time variation of the ionosphere, not only can high time delay prediction precision be maintained, but also the parameter model is unchanged within the same day time, a user only needs to obtain a parameter model once and measure a signal level in real time, and then the signal propagation time delay can be calculated, and the parameter model does not need to be frequently obtained. The short-wave time signal propagation time delay prediction method disclosed by the application is simple to realize, is helpful to engineering application implementation, can effectively improve short-wave time precision, and has positive significance for promoting development of a short-wave ionosphere monitoring and other related fields.

[0145] It should be noted that the above-described figures are only schematic representations of the processes included in the method according to the exemplary embodiments of the application, and are not intended to limit the purposes. It is easy to understand that the processes shown in the above-described figures do not indicate or limit the time sequence of the processes. In addition, it is also easy to understand that the processes can be executed synchronously or asynchronously, for example, in multiple modules.

[0146] Further, with reference to Figure 3 In the embodiment of the present example, a short-wave time signal propagation time delay prediction device 30 is also provided, comprising:

[0147] A time recurrence device 301 is configured to provide a reference 10MHz signal and a reference 1PPS for a short-wave time monitoring receiver;

[0148] A short-wave time monitoring receiver 302 is configured to receive a short-wave time signal, and perform timing, time difference measurement, and signal level measurement on the received short-wave time signal; and output time difference measurement data and signal level measurement data to an upper computer; wherein the time difference measurement data is used to convert a corresponding propagation time delay measurement value in the upper computer;

[0149] An upper computer 303 is configured to collect prior data of a target time length, divide and preprocess the prior data according to a preset rule to obtain sample data corresponding to each time period, and construct a sample data set; wherein the sample data of each time period includes a propagation time delay measurement value and a signal level measurement value of the short-wave time signal in the time period; and the sample data set is used to update parameters of a short-wave time signal propagation time delay prediction model to determine updated model hyperparameters; wherein the short-wave time signal propagation time delay prediction model is constructed based on a carrier frequency feature of the short-wave time signal, a signal level feature of the short-wave time signal at a current time, and a time period feature at the current time; and the model hyperparameters include hyperparameters corresponding to the signal level feature of the short-wave time signal at the current time and the time period feature at the current time;

[0150] The parameter injection unit 304 is configured to broadcast the updated model hyperparameters, so that the short-wave time service receiving terminal updates the short-wave time service signal propagation time delay prediction model according to the updated model hyperparameters, and performs the propagation time delay prediction of the short-wave time service signal by using the updated model.

[0151] For example, a short-wave time service monitoring station can be provided and deployed away from the short-wave time service broadcasting system, including a time recurrence device, a short-wave time service monitoring receiver, a host computer, and a data transmission unit. The time recurrence device is configured to provide the short-wave time service monitoring receiver with a reference 10MHz signal and a reference 1PPS required for operation. For example, the time recurrence device is configured to realize time and frequency synchronization with the BPM short-wave time service station in a satellite common view manner.

[0152] Reference Figure 4 As shown, the short-wave time service monitoring receiver is configured to receive the BPM short-wave time service signal and perform signal acquisition, timing, time difference measurement, and signal level measurement, and output the time difference measurement data and the signal level measurement data to the host computer. The timing refers to signal synchronization performed by the short-wave time service monitoring receiver after the BPM short-wave time service signal is acquired, and a local 1PPS is output, the front edge of the 1PPS is aligned with the starting time of the BPM short-wave time service signal. The time difference measurement refers to the measurement of the time difference between the reference 1PPS and the local 1PPS by the short-wave time service monitoring receiver, the time difference actually includes the fixed channel time delay of the short-wave time service monitoring receiver, and the unit is second. The signal level measurement refers to the measurement of the BPM short-wave time service signal level acquired by the short-wave time service monitoring receiver.

[0153] The host computer can be deployed at the short-wave time service monitoring station, and can include a data processing unit configured to construct a short-wave time service signal propagation time delay prediction model, and collect, store, and preprocess prior data, and train the time delay prediction model and update the model hyperparameters by using the prior data, so as to obtain optimal model parameters. The host computer can be connected to a data storage server, and the prior data received by the short-wave time service monitoring station can be stored in the data storage server. The host computer can also be connected to an industrial computer, and the industrial computer is configured to support the operation of the host computer. That is, the time delay prediction model is deployed at the short-wave time service monitoring station.

[0154] Specifically, the host computer can receive the time difference measurement data and the signal level measurement data output by the short-wave time service monitoring receiver, convert the time difference measurement data from seconds to microseconds, and then convert the signal propagation time delay measurement data by deducting the channel delay of the monitoring receiver. The channel delay of the monitoring receiver can be obtained by pre-calibration of the device.

[0155] The shortwave time synchronization monitoring station uses time delay data and signal level data to train the prediction model and obtain updated model hyperparameters for the current period. The updated model hyperparameters are then sent to the shortwave time synchronization broadcasting system via the data transmission unit, which broadcasts the updated model hyperparameters.

[0156] For example, refer to Figure 5 As shown, the data transmission unit may include: routers, VPNs, firewalls, network gateways, switches, log auditing components, internet access management components, bastion hosts, and other information technology equipment, forming a secure and reliable data transmission network for remote transmission of prior data. Specifically, routers enable network data access; VPNs establish dedicated communication lines using specially encrypted communication protocols; firewalls ensure the security of internal networks and computers; network gateways disconnect internal and external networks, enabling secure and appropriate application data exchange between networks; switches filter, classify, and forward data packets from different network segments; log auditing components store, monitor, audit, analyze, alarm, respond to, and report logs generated by various information technology devices; internet access management components prevent the malicious spread of illegal information and avoid data leakage, enabling real-time monitoring and management of network resource usage; and bastion hosts protect network data security from intrusion, preventing network and data from being compromised by other visitors.

[0157] Alternatively, in some exemplary embodiments, the host computer of the shortwave time synchronization monitoring station uses a data processing unit to preprocess the time delay data and signal level data of the shortwave time synchronization signal; then, it transmits the preprocessed data to the shortwave time synchronization broadcasting system via a data transmission unit. The shortwave time synchronization broadcasting system receives the time delay data and signal level data, and uses server equipment to train a time delay prediction model on the received data, calculates the updated model hyperparameters for the current period, and then broadcasts the data. That is, a time delay prediction model is deployed on one side of the shortwave time synchronization broadcasting system.

[0158] For example, a shortwave time synchronization broadcasting system may include a parameter injection unit for arranging the updated optimal parameter model into a message according to the format requirements of the shortwave time synchronization broadcasting system and injecting it into the shortwave time synchronization broadcasting system for broadcasting to users.

[0159] For example, updated model hyperparameters can be embedded in the shortwave time signal message. For instance, the model hyperparameters can be compiled and written into a reserved field or an idle field of the message. Upon receiving the message, the shortwave time receiver can parse it and obtain the corresponding model hyperparameters.

[0160] It should be noted that, although several modules or units of the devices for action execution are mentioned in the above detailed description, such a division is not mandatory. Indeed, according to an embodiment of the application, the features and functionalities of two or more modules or units described above can be embodied in one module or unit. Conversely, the features and functionalities of one module or unit described above can be further divided into embodied by a plurality of modules or units.

[0161] Figure 6 A schematic diagram of an electronic device suitable for implementing embodiments of the application is shown.

[0162] It should be noted that, Figure 6 The electronic device 1000 shown is merely an example and should not limit the function and scope of use of embodiments of the application in any way.

[0163] As Figure 6 shown, the electronic device 1000 includes a central processing unit (CPU) 1001, which can perform various appropriate actions and processes in accordance with a program stored in a read-only memory (ROM) 1002 or a program loaded from a storage section 1008 into a random access memory (RAM) 1003. Various programs and data required for system operation are also stored in the RAM 1003. The CPU 1001, the ROM 1002, and the RAM 1003 are connected to each other through a bus 1004. An input / output (I / O) interface 1005 is also connected to the bus 1004. Further, the electronic device 1000 includes an FPGA device, a SOC device.

[0164] The following components are connected to the I / O interface 1005: an input section 1006 including a keyboard, a mouse, etc.; an output section 1007 including a display such as a cathode ray tube (CRT), a liquid crystal display (LCD), etc., and a speaker, etc.; a storage section 1008 including a hard disk, etc.; and a communication section 1009 including a network interface card such as a LAN (Local Area Network) card, a modem, etc. The communication section 1009 performs communication processing via a network such as the Internet. A drive 1010 is also connected to the I / O interface 1005 as necessary. A removable recording medium 1011 such as a magnetic disk, an optical disk, a magneto-optical disk, a semiconductor memory, etc. is attached to the drive 1010 as necessary, so that a computer program read out therefrom is installed in the storage section 1008 as necessary.

[0165] For example, the electronic device described above can be a host computer.

[0166] In particular, according to embodiments of the present application, the processes described below with reference to the flowcharts can be implemented as a computer software program. For example, embodiments of the present application include a computer program product comprising a computer program carried on a storage medium, the computer program comprising program code for performing the methods illustrated by the flowcharts. In such embodiments, the computer program can be downloaded and installed from a network by the communication section 1009, and / or installed from the detachable medium 1011. When the computer program is executed by the central processing unit (CPU) 1001, various functions defined in the system of the present application are performed.

[0167] In particular, the electronic device described above can be an onboard intelligent electronic device, such as an onboard video processing device.

[0168] It should be noted that the storage medium shown in the embodiments of the present application can be a computer readable signal medium or a computer readable storage medium or any combination of the two. The computer readable storage medium may, for example, but is not limited to, an electrical, magnetic, optical, electromagnetic, infrared, or semiconductor system, device or component, or any combination of the above. More specific examples of the computer readable storage medium can include, but are not limited to, an electrical connection having one or more wires, a portable computer diskette, a hard disk, a random access memory (RAM), a read-only memory (ROM), an erasable programmable read-only memory (EPROM), a flash memory, an optical fiber, a portable compact disk read-only memory (Compact Disc Read-Only Memory, CD-ROM), an optical storage device, a magnetic storage device, or any suitable combination of the above. In the present application, the computer readable storage medium can be any tangible medium containing or storing a program that can be used by or in conjunction with an instruction execution system, device or component. In the present application, the computer readable signal medium can include a data signal carried in a baseband or as a part of a carrier wave, which carries computer readable program code. Such a propagated data signal can take many forms, including but not limited to an electromagnetic signal, an optical signal or any suitable combination of the above. The computer readable signal medium can also be any storage medium other than the computer readable storage medium that can send, propagate or transmit a program for use by or in conjunction with an instruction execution system, device or component. The program code contained on the storage medium can be transmitted by any suitable medium, including but not limited to wireless, wired, or the like, or any suitable combination of the above.

[0169] The computer program product of the present application includes a computer program, which, when executed by a processor, implements the steps of the above-mentioned method embodiments.

[0170] The units described in the embodiments of the present application can be implemented by software, or by hardware, or by a combination of software and hardware. The units described may also be implemented as a processor. In some cases, the names of the units are not intended to be limiting.

[0171] It should be noted that, as another aspect, the present application also provides a storage medium, which can be included in an electronic device, or can exist independently without being assembled into the electronic device. The storage medium carries one or more programs, which, when executed by an electronic device, cause the electronic device to implement the method described in the above embodiments. For example, the electronic device can implement each step of the method shown in Figure 1

[0172] In one embodiment, the present application provides a computer program product, which includes a computer program that, when executed by a processor, implements the steps of the above-mentioned method embodiments.

[0173] In addition, the above-mentioned drawings are only schematic illustrations of the processes included in the method according to the exemplary embodiments of the present application, and are not intended to be limiting. It is easy to understand that the processes shown in the above-mentioned drawings do not indicate or limit the time sequence of the processes. In addition, it is also easy to understand that the processes can be executed synchronously or asynchronously, for example, in multiple modules.

[0174] ​Other embodiments of the application will be apparent to those skilled in the art from consideration of the specification and practice of the application disclosed herein. It is intended that the specification and examples be considered as exemplary only, with a true scope and spirit of the application being indicated by the following claims.

[0175] It is to be understood that the application is not limited to the precise structures hereinabove described and shown in the drawings, for purposes of illustration and education only, and that variations and changes can be made by persons skilled in the art in the application without departing from the scope of the application. The scope of the application is limited only by the claims that follow.

Claims

1. A short wave time signal propagation delay prediction method, characterized in that, The method comprises: Collecting prior data of a target duration, performing time period division and preprocessing on the prior data according to a preset rule to obtain sample data corresponding to each time period, and constructing a sample data set; wherein the sample data of each time period comprises the propagation delay of the short-wave time signal in the time period and the signal level; Using the sample data set to update the parameters of the short-wave time signal propagation delay prediction model to determine the updated model hyperparameters; wherein the model hyperparameters comprise the signal level feature of the current short-wave time signal and the hyperparameters corresponding to the time period feature of the current time; the short-wave time signal propagation delay prediction model is constructed based on the carrier frequency feature of the short-wave time signal, the signal level feature of the current short-wave time signal, and the time period feature of the current time, and comprises: wherein, is a date serial number d , t is a signal propagation delay prediction value of the shortwave time signal at the time instant; is a date serial number d , t is a signal level of the shortwave time signal at the time instant, used to represent a signal level feature of the shortwave time signal at the current time instant; is a time period corresponding to the time instant t , used to represent a time period feature corresponding to the current time instant; represents a carrier frequency feature of the shortwave time signal; , , are model hyperparameters, respectively; Broadcasting the updated model hyperparameters for the short-wave time signal receiving terminal to update the short-wave time signal propagation delay prediction model according to the updated model hyperparameters, and using the updated model to predict the propagation delay of the short-wave time signal.

2. The method of claim 1, wherein, The method further comprises: constructing the short-wave time signal propagation delay prediction model in advance according to the carrier frequency feature of the short-wave time signal, the signal level feature of the current short-wave time signal, and the time period feature corresponding to the current time.

3. The method of claim 1, wherein, The method further comprises: configuring a carrier frequency feature of the short-wave time signal according to a carrier frequency of the short-wave time signal comprising: wherein f denotes the carrier frequency of the short wave time signal.

4. The method of claim 1, wherein, Collecting prior data of a target duration, performing time period division and preprocessing on the prior data according to a preset rule to obtain sample data corresponding to each time period, comprising: Collecting the signal propagation delay measurement value of the short-wave time signal and the signal level measurement value of the short-wave time signal corresponding to a plurality of dates before the current date; Sort data by date and configure corresponding date serial number d ; Dividing each day into M time periods and configuring a corresponding time period number λ for each time period; Calculating the arithmetic mean value of the signal propagation delay measurement value in each time period and configuring it as the signal propagation delay corresponding to the time period; Calculating the arithmetic mean value of the signal level measurement value in each time period and configuring it as the signal level corresponding to the time period.

5. The method of claim 4, wherein, The method further comprises: configuring a value range of a time period number M as: , and being divisible by 86400; wherein, .

6. The method of claim 1, wherein, Using the sample data set to update the parameters of the short-wave time signal propagation delay prediction model to determine the updated model hyperparameters, comprising: Dividing the sample data set into a training set and a validation set; defining a model parameter vector from model hyperparameters ; wherein ; The time delay variable is defined according to the signal time delay of each time period, and a signal delay feature is constructed in combination of the time delay variable and the carrier frequency feature ; and a signal delay feature vector is configured according to the signal delay feature ; wherein, , ; represents the carrier frequency feature of the short-wave time signal. The signal level variable is defined according to the signal level of each time period, and a signal level matrix is constructed in combination of the signal level variable and the time period characteristics ; wherein, wherein denotes a time period, M denotes the maximum time period value; is t the average value of the signal level of the short-wave time signal within the time period; using the signal level matrix corresponding to the training set , a signal delay feature vector , determining a model parameter vector , comprising: ; The model parameter vector is substituted into the delay prediction model to obtain a corresponding propagation delay prediction value; an error between the delay prediction value and the delay measurement value is calculated; verifying set to the model parameter vector corresponding to the minimum mean absolute error is configured as the updated model hyperparameter.

7. The method of claim 6, wherein, Configuring the model parameter vector corresponding to the minimum mean absolute error as the updated model hyperparameters, comprising: wherein, represents when is a minimum, the model parameter vector employed in calculating the signal propagation delay prediction value ; is a signal propagation delay prediction value for the day number , the time period; is a signal propagation delay for the day number , the time period.

8. The method of claim 1, wherein, Using the updated model to predict the propagation delay of the short-wave time signal, comprising: Inputting the short-wave signal level corresponding to the current short-wave time signal and the time period corresponding to the current short-wave time signal into the updated model to obtain the short-wave time signal propagation delay of the current short-wave time signal output by the model.

9. A short wave time signal propagation delay prediction device, characterized by, The device comprises: A time reproduction device for providing a reference 10MHz signal and a reference 1PPS for a short-wave time monitoring receiver; A short-wave time monitoring receiver for receiving a short-wave time signal and performing timing, time difference measurement, and signal level measurement on the received short-wave time signal; and outputting the time difference measurement data and the signal level measurement data to an upper computer; wherein the time difference measurement data is used to convert the corresponding propagation delay measurement value in the upper computer; The host computer is configured to collect prior data of a target time length, divide and preprocess the prior data according to a preset rule to obtain sample data corresponding to each time period, and construct a sample data set; wherein the sample data of each time period includes the propagation delay of the short-wave time signal in the time period and the signal level; the sample data set is used to update the parameters of the short-wave time signal propagation delay prediction model to determine the updated model hyperparameters; wherein the model hyperparameters include the signal level feature of the short-wave time signal at the current time and the hyperparameters corresponding to the time period feature at the current time; the short-wave time signal propagation delay prediction model is constructed based on the carrier frequency feature of the short-wave time signal, the signal level feature of the short-wave time signal at the current time, and the time period feature at the current time, and includes: wherein, is a date serial number d , t is a signal propagation delay prediction value of the time instant; is a date serial number d , t is a signal level of the time instant, used to represent the signal level feature of the time instant; is a time period corresponding to the time instant, used to represent the time period feature corresponding to the time instant; t is a time period corresponding to the time instant, used to represent the time period feature corresponding to the time instant; represents the carrier frequency feature of the short wave time signal; , , are model hyperparameters, respectively; The parameter injection unit is configured to broadcast the updated model hyperparameters, so that the short-wave time signal receiving terminal updates the short-wave time signal propagation delay prediction model according to the updated model hyperparameters, and uses the updated model to predict the propagation delay of the short-wave time signal.

10. A computer program product comprising a computer program, characterized in that, The computer program is executed by the processor to implement the short-wave time signal propagation delay prediction method of any one of claims 1-8.

Images