Method, system, electronic device, program product and storage medium for predicting a loran time delay signal

By combining a multi-period term and trend term model with wavelet denoising technology, the problem of insufficient prediction accuracy of Rowland time delay signals was solved, achieving high-precision prediction of Rowland time delay signals in complex electromagnetic environments and improving the accuracy of navigation and positioning.

CN121500349BActive Publication Date: 2026-03-27NAT TIME SERVICE CENT CHINESE ACAD OF SCI
View PDF 2 Cites 0 Cited by

Patent Information

Application Number
CN202610035011.6
Authority / Receiving Office
CN · China
Patent Type
Patents(China)
Current Assignee / Owner
Filing Date
2026-01-12
Publication Date
2026-03-27
Estimated Expiration
2046-01-12

AI Technical Summary

Technical Problem

Existing Roland time delay signal prediction methods are complex in their formation mechanism, making it difficult to effectively capture multi-scale periodic, non-stationary disturbances and long-term trends, resulting in insufficient prediction accuracy, especially poor positioning performance in complex electromagnetic environments.

Method used

By combining a multi-period and trend term model with wavelet denoising technology, periodic and trend terms are extracted through FFT spectral analysis. The residual signal is processed using Symlet4 wavelet decomposition and threshold denoising. Finally, the multi-period, trend, and residual signals are fused through an adaptive weighting mechanism to achieve accurate prediction.

Benefits of technology

It significantly improves the prediction accuracy of Roland time delay signals, especially enabling high-precision navigation and positioning in complex electromagnetic environments, adapting to long-term trends and parameter drift, and improving the accuracy of navigation and positioning.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure CN121500349B_ABST
    Figure CN121500349B_ABST
Patent Text Reader

Abstract

The application provides a kind of Loran time delay signal prediction method, system, electronic equipment, program product and storage medium, the method comprises: obtaining initial Loran time delay signal;Periodic term of the initial Loran time delay signal is extracted, and input to the multi-periodic term and trend term model constructed;Residual signal between the observation value of initial Loran time delay signal and the output of the multi-periodic term and trend term model is calculated;Wavelet decomposition and threshold denoising are carried out on residual signal;The output of the multi-periodic term and trend term model and the denoised residual signal are fused by adaptive weight mechanism to obtain predicted Loran time delay signal;The certainty periodic term and long-term trend drift of Loran time delay signal are captured by multi-periodic term and trend term model, and non-stationary disturbance is handled using wavelet denoising, which significantly improves the prediction accuracy of Loran time delay, especially suitable for high-precision timing scene in complex electromagnetic environment.
Need to check novelty before this filing date? Find Prior Art

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of radio navigation positioning, and in particular to a Loran time delay signal prediction method, system, electronic device, program product and storage medium. BACKGROUND

[0002] As an important land-based radio navigation positioning system, the Loran system has significant advantages such as stable signal propagation, wide coverage, and strong anti-interference capability. In scenarios where Global Navigation Satellite System (GNSS) signals are blocked or fail (such as high-rise dense areas, extreme weather conditions, and electromagnetic interference environments), the Loran system can provide reliable backup navigation and positioning services and is an important complementary means for GNSS. Its core principle is to measure the time difference (time delay) of signals from different transmitting stations to the receiver to achieve positioning, so the time delay prediction accuracy directly determines the positioning performance of the system.

[0003] The existing Loran time delay signal formation mechanism is complex, which leads to insufficient prediction of the existing Loran time delay signal. The Loran time delay signal is mainly affected by the following factors:

[0004] Multi-scale periodicity: diurnal cycle, diurnal variation of ionospheric equivalent reflection height causes time delay to present 24-hour periodic fluctuation; seasonal cycle, seasonal changes in weather conditions (such as temperature, humidity, atmospheric pressure) caused by seasonal changes cause time delay to have seasonal periodic characteristics; interannual cycle: interannual variation of solar activity (such as sunspot cycle) will cause slow drift of periodic parameters (amplitude, phase).

[0005] Non-stationary disturbance: local terrain attenuation, signal propagation loss differences of different terrains will introduce non-stationary noise; sudden weather disturbance, such as thunderstorms, strong winds, etc. Extreme weather can cause short-term mutations in time delay; measurement noise, receiver hardware noise, clock jitter, etc. Random factors will be superimposed in the time delay signal.

[0006] Long-term trend: crustal movement, the movement of the earth's plates will cause the geometric distance between the transmitting station and the receiver to change slowly, which is reflected as a long-term linear trend in time delay; device aging, aging of the transmitting station and receiver devices will cause slow drift of signal propagation characteristics. SUMMARY

[0007] The present application proposes a Loran time delay signal prediction method, system, electronic device, program product and storage medium to solve the problem of insufficient prediction of the existing Loran time delay signal caused by the complex formation mechanism of the existing Loran time delay signal in the prior art.

[0008] In a first aspect, the present application embodiment provides a Loran time delay signal prediction method, which comprises:

[0009] Obtaining an initial Loran time delay signal;

[0010] extracting the periodic terms of the initial Loran time delay signal and inputting into a constructed multi-periodic and trend term model; the multi-periodic and trend term model is:

[0011]

[0012] wherein, is the frequency of the ith periodic term, t is the current time, N is the number of periodic terms, is the sine amplitude of the ith periodic term, is the cosine amplitude of the ith periodic term, is the trend slope, and is a constant term;

[0013] calculating a residual signal between the observation value of the initial Loran time delay signal and the output of the multi-periodic and trend term model; the residual signal is:

[0014]

[0015] wherein, is the observation value of the initial Loran time delay signal, is the output of the multi-periodic and trend term model;

[0016] wavelet decomposing and threshold denoising the residual signal; the denoised residual signal is:

[0017]

[0018] wherein, J is the decomposition level of the wavelet decomposition, is the Jth layer approximation coefficient, is the jth layer detail coefficient after denoising;

[0019] fusing the output of the multi-periodic and trend term model and the denoised residual signal through an adaptive weight mechanism to obtain a predicted Loran time delay signal; the predicted Loran time delay signal is:

[0020]

[0021] wherein, ∈[0,1] is the adaptive weight.

[0022] In one embodiment, before the extracting the periodic terms of the initial Loran time delay signal and inputting into a constructed multi-periodic and trend term model, it further comprises:

[0023] The initial Loran time delay signal is subjected to a de-meaning process; the initial Loran time delay signal after the de-meaning process is:

[0024]

[0025] The initial Loran time delay signal is , M is the number of sampling points, is a sampling interval, is the kth sampling point, is the observation value corresponding to the kth sampling point, is the mean value of the initial Loran time delay signal of all sampling points.

[0026] In one embodiment, the wavelet decomposition and threshold denoising of the residual signal comprises:

[0027] Symlet4 wavelet is used as the base function of wavelet decomposition;

[0028] In one embodiment, the decomposition layer number J is:

[0029]

[0030] M is the number of sampling points of the initial Loran time delay signal.

[0031] In one embodiment, the wavelet decomposition and threshold denoising of the residual signal comprises:

[0032] The threshold method is used to denoise the detail coefficient , and the threshold value is:

[0033]

[0034] , where is the standard deviation of the noise of the jth layer, , is the median absolute deviation of ;

[0035] The soft threshold value processing is performed on the detail coefficient , and the formula is:

[0036]

[0037] , where is the jth layer detail coefficient after soft threshold value processing, and k is the coefficient index.

[0038] In one embodiment, the adaptive weight is: ​​

[0039]

[0040] wherein, is the maximum weight, and γ is the adaptation rate, is the average absolute error of the last W sampling points, is the historical average error.

[0041] In one embodiment, the multi-periodic term and trend term model updates the trend term through a time window; the update formula of the trend term is:

[0042]

[0043] wherein, is the updated trend slope, is the updated constant term, is the time window, is the kth sampling point of the initial Loran time delay signal.

[0044] In a second aspect, the embodiments of the present application provide a prediction system of a Loran time delay signal, the system comprising:

[0045] an acquisition module, configured to acquire an initial Loran time delay signal;

[0046] an extraction module, configured to extract a periodic term of the initial Loran time delay signal and input the periodic term into a constructed multi-periodic term and trend term model; the multi-periodic term and trend term model is:

[0047]

[0048] wherein, is the frequency of the ith periodic term, t is the current time, and N is the number of periodic terms, is the sine amplitude of the ith periodic term, is the cosine amplitude of the ith periodic term, is the trend slope, and is the constant term;

[0049] a residual calculation module, configured to calculate a residual signal between an observation value of the initial Loran time delay signal and an output of the multi-periodic term and trend term model; the residual signal is:

[0050]

[0051] wherein, is the observation value of the initial Loran time delay signal, is the output of the multi-periodic term and trend term model;

[0052] a wavelet denoising module, configured to perform wavelet decomposition and threshold denoising on the residual signal; the denoised residual signal is:

[0053]

[0054] wherein J is a decomposition level of the wavelet decomposition, is an approximate coefficient of the Jth level, is a denoised detail coefficient of the jth level;

[0055] a fusion module, configured to fuse the output of the multi-period term and the trend term model and the denoised residual signal through an adaptive weight mechanism to obtain a predicted Loran delay signal; the predicted Loran delay signal is:

[0056]

[0057] wherein ∈ [0, 1] is an adaptive weight.

[0058] In an embodiment, the system further comprises:

[0059] a mean removal module, configured to perform mean removal processing on the initial Loran delay signal; the mean removal processed initial Loran delay signal is:

[0060]

[0061] wherein the initial Loran delay signal is , M is a number of sampling points, is a sampling interval, is a kth sampling point, is an observation value corresponding to the kth sampling point, is a mean value of the initial Loran delay signal of all sampling points.

[0062] In an embodiment, the wavelet denoising module is further configured to:

[0063] adopt a Symlet4 wavelet as a base function of the wavelet decomposition;

[0064] In an embodiment, the decomposition level J is:

[0065] ;

[0066] wherein M is a number of sampling points of the initial Loran delay signal.

[0067] In an embodiment, the wavelet denoising module is further configured to:

[0068] The threshold method is used to denoise the detail coefficients

[0069]

[0070] wherein, is the standard deviation of the noise of the jth layer, , is the median absolute deviation of the jth layer, ;

[0071] The soft threshold processing is performed on the detail coefficients The formula is:

[0072]

[0073] wherein, is the jth layer detail coefficient after the soft threshold processing, and k is the coefficient index.

[0074] In an embodiment, the adaptive weight

[0075] wherein,

[0076] is the maximum weight, γ is the adaptive rate, is the average absolute error of the last W sampling points, is the historical average error. In an embodiment, the multi-period term and the trend term model update the trend term through a time window; the update formula of the trend term is:

[0077]

[0078] wherein,

[0079] is the updated trend slope, is the updated constant term, is the time window, is the kth sampling point of the initial Loran time delay signal. In a third aspect, an embodiment of the present application provides an electronic device, which comprises a processor and a memory, and the memory stores at least one instruction, at least one program, a code set or an instruction set, and the at least one instruction, the at least one program, the code set or the instruction set is loaded and executed by the processor to realize the prediction method of the Loran time delay signal as described in any one of the first aspect.

[0080]

[0081] ​​​​In a fourth aspect, an embodiment of the present application provides a computer readable storage medium, and the computer readable storage medium stores at least one instruction or at least one program, and the at least one instruction or the at least one program is loaded and executed by a processor to implement the prediction method of the Loran time delay signal according to any one of the first aspect.

[0082] In a fifth aspect, an embodiment of the present application provides a computer program product, and the computer program product comprises a computer program, and the computer program is executed by a processor to implement the prediction method of the Loran time delay signal according to any one of the first aspect.

[0083] The prediction method of the Loran time delay signal, the system, the electronic device, the program product and the storage medium provided by the present application have the following technical effects:

[0084] The deterministic periodic term and the long-term trend drift of the Loran time delay signal are captured through the multi-period term and the trend term model, and the non-stationary disturbance is processed by using the wavelet denoising. Finally, the adaptive weight fusion is performed, the adaptability of the model to the long-term trend and the parameter drift is significantly improved, the prediction accuracy of the Loran time delay is significantly improved, and the method is especially suitable for high-precision timing scenes in a complex electromagnetic environment.

[0085] 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

[0086] The drawings constituting a part of the present application are used to provide further understanding of the present application, and the schematic embodiments of the present application and the description thereof are used to explain the present application, and do not constitute an improper limitation on the present application. In the drawings:

[0087] Figure 1 A flowchart of a prediction method of a Loran time delay signal is provided for an exemplary embodiment of the present application;

[0088] Figure 2 A schematic diagram of an initial Loran time delay signal is provided for an exemplary embodiment of the present application;

[0089] Figure 3 A schematic diagram of a spectrum analysis of an initial Loran time delay signal is provided for an exemplary embodiment of the present application;

[0090] Figure 4 An output schematic diagram of a multi-period term and a trend term model is provided for an exemplary embodiment of the present application;

[0091] Figure 5 A prediction Loran time delay signal obtained by fusing the output of the multi-period term and the trend term model and the denoised residual signal is provided for an exemplary embodiment of the present application;

[0092] Figure 6 The joint schematic diagram of the initial Loran time delay signal, the output of the multi-period term and the trend term model, the output of the multi-period term and the trend term model, and the predicted Loran time delay signal obtained by fusing the denoised residual signal provided by an exemplary embodiment of the present application;

[0093] Figure 7 The structural schematic diagram of a Loran time delay signal prediction system provided by an exemplary embodiment of the present application;

[0094] Figure 8 The structural schematic diagram of an electronic device provided by an exemplary embodiment of the present application. DETAILED DESCRIPTION

[0095] The present application will be described in detail below with reference to the accompanying drawings and in conjunction with embodiments. It should be noted that the embodiments in the present application and the features in the embodiments can be combined with each other without conflict.

[0096] The following detailed description is exemplary description, which is intended to provide further detailed description of the present application. Unless otherwise specified, all technical terms used in the present application have the same meaning as generally understood by those skilled in the art to which the present application belongs. The terms used in the present application are only for the purpose of describing the specific embodiments, and are not intended to limit the exemplary embodiments according to the present application.

[0097] At present, the prediction methods for Loran time delay signals mainly include the following three types:

[0098] The first type is a method based on a physical model.

[0099] The time delay prediction is realized by establishing a physical mechanism model of Loran time delay signal propagation (such as Norton model of ground wave propagation and ionospheric reflection model of sky wave propagation). Its limitation lies in that the model parameters (such as ionospheric equivalent height and terrain attenuation coefficient) are difficult to accurately obtain, and the calculation complexity is high, which is difficult to meet the real-time prediction requirement.

[0100] The second type is a method based on a statistical model.

[0101] Typical representatives include Fourier series model, ARIMA (autoregressive integrated moving average) model, etc. Its limitation lies in that the Fourier series model has weak processing ability for non-stationary disturbance; the ARIMA model is difficult to depict the coupling relationship of multi-scale periods.

[0102] The third type is a method based on data-driven.

[0103] Such as neural network, support vector machine, etc., realize prediction by learning nonlinear mode in data. Its limitation lies in dependence on large amount of labeled data, and lack of utilization of periodic priori knowledge, and precision obviously decreases in data sparse or periodic parameter drift scene.

[0104] For the specific problems in the above scheme, an example embodiment of the present application provides a method for predicting Loran time delay signal, referring to Figure 1 , the method comprises:

[0105] S101, obtaining an initial Loran time delay signal.

[0106] In an embodiment, in order to eliminate the influence of the direct current component in the initial Loran time delay signal, the initial Loran time delay signal is subjected to mean removal processing, and after step S101, the method further comprises:

[0107] The initial Loran time delay signal is subjected to mean removal processing; the initial Loran time delay signal subjected to mean removal processing is:

[0108]

[0109] Wherein, the initial Loran time delay signal is , M is the number of sampling points, is the sampling interval, is the kth sampling point, is the observation value corresponding to the kth sampling point in the initial Loran time delay signal, is the mean value of the initial Loran time delay signal of all sampling points in the initial Loran time delay signal.

[0110] S102, extracting a periodic term of the initial Loran time delay signal and inputting to a constructed multi-periodic term and trend term model.

[0111] Specifically, the extracting the periodic term of the initial Loran time delay signal comprises: performing periodic detection on the initial Loran time delay signal through FFT spectrum analysis to obtain the number N of periodic terms and the specific value of the periodic term.

[0112] The multi-periodic term and trend term model is:

[0113]

[0114] Wherein, is the frequency of the i th periodic term, t is the current time, N is the number of periodic terms, is the sine amplitude of the i th periodic term, is the cosine amplitude of the i th periodic term, is the trend slope, and is a constant term. ​​And N are obtained by FFT spectrum analysis of the initial Loran time delay signal.

[0115] Specifically, the core features of the Loran time delay signal can be summarized as the superposition of periodic terms, long-term trends and non-stationary disturbances. In this embodiment, the Loran time delay signal is divided into:

[0116]

[0117] Among them, is the periodic term, caused by regular factors such as earth rotation and solar radiation, including diurnal (24h), seasonal (365d) and other multi-scale periods; is the trend term, reflecting the slow change of the Loran time delay signal at the interannual or longer time scale, such as the slow increase of the distance between the transmitting station and the receiver caused by the crustal movement; is the non-stationary disturbance term, caused by sudden ionospheric anomalies, local weather fluctuations, measurement noise, etc., and characterized by random or short-time burst characteristics. In this embodiment, the periodic term and the trend term are mainly analyzed.

[0118] Regarding the periodic term, assuming that the Loran time delay signal contains N main periodic terms, the periodic term can be expressed as:

[0119]

[0120] Among them, is the frequency (unit: Hz) of the i-th periodic term, is the period (unit: s); is the sine amplitude of the i-th periodic term, is the cosine amplitude of the i-th periodic term. Specifically, the main periodic terms can include: diurnal period T1=24×3600=86400s (ionospheric diurnal variation caused by earth rotation); semi-diurnal period T2=12×3600=43200s (ionospheric semi-diurnal fluctuation); seasonal period T3=365×86400=3.1536×107s (seasonal change of solar elevation angle). Generally, the number of periodic terms and the observed values of the periodic terms can be obtained by FFT spectrum analysis of the initial Loran time delay signal.

[0121] Specifically, the amplitude parameters

[0122] , can be determined by least squares fitting. Let the observed value of a certain sampling point in the initial Loran signal be , and define the error function as:

[0123] ​

[0124] To minimize the error function For amplitude parameters , Taking the partial derivatives and setting them to zero, we obtain a system of linear equations:

[0125]

[0126] The amplitude parameters for each period can be obtained by solving the above set of equations. , The estimated value.

[0127] Regarding the trend term, it is used to capture long-term, slow changes in time-delayed signals. The mathematical expression for the trend term is:

[0128]

[0129] in, The trend slope (unit: s / s). This is a constant term (unit: s).

[0130] Specifically, trend parameters , The error function is defined as follows: (Determined by least squares fitting)

[0131]

[0132] To minimize the error function ,right , Taking the partial derivatives and setting them to zero, we obtain a system of linear equations:

[0133]

[0134] The estimated values ​​of k and b can be obtained by solving the above system of equations. , The trend term used to fit the Rowland time delay signal .

[0135] In this embodiment, a multi-period and trend term model is constructed through FFT spectrum analysis to accurately extract the periodic term in the time-delay signal, and a linear trend term is introduced to capture long-term change patterns.

[0136] In one embodiment, to accommodate the slow drift of the periodic parameter, the multi-period term and trend term model updates the trend term through a time window; the update formula for the trend term is:

[0137]

[0138] in, is a periodic term, is the updated trend slope, is the updated constant term, is the time window, is the kth sample point of the initial Loran time delay signal. Generally, the time window can be set as i.e. one hour, the time window can be a sliding time window.

[0139] In an alternative embodiment, regarding the update strategy of the model, to reduce the amount of calculation, the recursive least square (RLS) method can be used to realize the online update of the trend term, avoiding the full amount of re-fitting each time;

[0140] In an alternative embodiment, regarding the update strategy of the model, a linear transition is used between the new trend term and the old trend term, avoiding the prediction jump caused by the sudden change of the trend term:

[0141]

[0142] wherein β = min(0.11·Δt, 1) is the transition coefficient, and Δt is the update time interval.

[0143] S103, calculate the residual signal between the observation value of the initial Loran time delay signal and the output of the multi-periodic term and trend term model.

[0144] The residual signal is:

[0145]

[0146] wherein, is the observation value of the initial Loran time delay signal, is the output of the multi-periodic term and trend term model. The residual signal contains fine-scale useful signals (such as small fluctuations caused by local weather) and noise (such as measurement error), so it is necessary to extract useful information by wavelet analysis for denoising the residual.

[0147] S104, wavelet decomposition and threshold denoising are performed on the residual signal.

[0148] The denoised residual signal is:

[0149]

[0150] wherein J is the decomposition level of the wavelet decomposition, is the Jth layer approximation coefficient, is the jth layer detail coefficient after denoising.

[0151] Specifically, the wavelet decomposition process can be regarded as projecting the signal onto a series of wavelet basis functions of different scales, thereby obtaining the signal details at different scales. Through multi-level decomposition, the residual signal can be decomposed into multiple levels of approximation coefficients and detail coefficients. The approximation coefficients correspond to the low-frequency components of the residual signal, reflecting the overall trend of the signal; the detail coefficients correspond to the high-frequency components of the residual signal, reflecting the local changes and noise of the residual signal.

[0152] In one embodiment, step S104 uses Symlet4 wavelet as the basis function of wavelet decomposition. The Symlet4 (sym4) wavelet is suitable for non-stationary signal processing due to its good symmetry and compact support characteristics. The Symlet wavelet is an improved version of the Daubechies wavelet, which further optimizes the symmetry while maintaining the compact support characteristics, reducing the phase distortion during signal reconstruction, and is particularly suitable for scenarios such as time delay signals that require accurate phase information.

[0153] The number of decomposition levels J is:

[0154]

[0155] where M is the number of sampling points of the initial Loran time delay signal, is a floor function, and J≤5 is limited to avoid information loss caused by excessive decomposition. Too many decomposition levels will cause the signal to be oversmoothed, losing detail information, while too few decomposition levels will not effectively separate noise. Through experimental verification, J=5 can achieve the best balance between denoising effect and computational efficiency.

[0156] Step S104 uses a threshold method to denoise the detail coefficients .

[0157] Specifically, the SureShrink threshold method can be used to denoise the detail coefficients dj. The SureShrink threshold method is an adaptive threshold selection method based on Stein's unbiased risk estimate (SURE), which can achieve the best balance between denoising and signal fidelity. For the jth layer detail coefficient, the threshold is:

[0158]

[0159] where is the standard deviation of the jth layer noise, , is the median absolute deviation of . .

[0160] The detail coefficients are soft-thresholded; the formula is:

[0161]

[0162] wherein, is the jth layer detail coefficient after soft thresholding, k is the coefficient index. Soft thresholding can remove noise while preserving the main features of the signal, avoiding signal oscillation that may be caused by hard thresholding.

[0163] Wavelet reconstruction of the residual signal after soft thresholding gives:

[0164]

[0165] wherein, is the denoised residual signal, J is the number of decomposition layers of wavelet decomposition, is the Jth layer approximation coefficient, is the jth layer detail coefficient after denoising, wherein the useful signal of the fine scale is effectively preserved, and the high-frequency noise is significantly suppressed.

[0166] S105, the outputs of the multi-periodic term and the trend term model and the denoised residual signal are fused through an adaptive weight mechanism to obtain a predicted Loran delay signal.

[0167] wherein, the predicted Loran delay signal is used to know the trend of the change of signal propagation delay at future time in advance, so as to correct the navigation and timing calculation. For example, navigation positioning error correction, timing synchronization accuracy improvement, etc., which will not be described here.

[0168] Since the multi-periodic parameters (such as amplitude , ) will slowly drift over time (such as interannual scale solar activity changes), the prediction accuracy of the model will decrease. Therefore, an adaptive weight mechanism is designed to dynamically fuse the outputs of the multi-periodic term and the trend term model and the residual signal after wavelet denoising to obtain a predicted Loran delay signal is:

[0169]

[0170] wherein, ∈[0,1] is an adaptive weight, reflecting the credibility of the periodic model: The closer to 1, the higher the proportion of the periodic model; The closer to 0, the stronger the role of residual correction.

[0171] Specifically, the adaptive weight is based on the ratio of the recent prediction error and the historical average error, and the specific formula is:

[0172]

[0173] where, is the maximum weight, and γ is the adaptation rate, is the mean absolute error of the last W sampling points, is the historical mean error.

[0174] The calculation formula of the mean absolute error is:

[0175]

[0176] The calculation formula of the historical mean error is:

[0177]

[0178] In order to avoid excessive fluctuations in the weight, the adaptive weight is subjected to a boundary constraint:

[0179]

[0180] where, is the minimum weight, which ensures that the model always maintains a certain contribution and prevents excessive residual correction from causing the model to be unstable.

[0181] In some alternative embodiments, in a stationary section with small prediction error (such as the stable phase of the diurnal cycle), is maintained between 0.8-1.0, and the model dominates; in a non-stationary section with increased error (such as a sudden ionospheric disturbance), is reduced to 0.3-0.5, and the correction effect of the residual signal is enhanced; the weight changes smoothly without sharp fluctuations, indicating that the adaptation rate setting of γ=0.05 is reasonable.

[0182] The following describes the present embodiment through a specific embodiment:

[0183] A certain Loran-C test station 48-hour continuous initial Loran time delay signal (sampling rate 1 Hz, total 172,800 data points) is obtained, as shown in Figure 2 , which contains complete diurnal cycle changes. The first 24-hour data (86,400 points) is used as the training set, and the last 24-hour data (86,400 points) is used as the test set.

[0184] First, the initial Loran time delay signal is de-meaned, and the mean value of the training set is calculated , the whole data is de-meaned, and the following is obtained:

[0185]

[0186] Next, the multi-period term and trend term model is constructed. Regarding the period term, FFT spectrum analysis is used, as shown inFigure 3 As shown, the frequency is obtained. That is, the day-night cycle The amplitude parameters were determined by least squares fitting, thus obtaining the amplitude parameters. , , , , , The trend slope is obtained by fitting the residual signal of the training set. constant term .

[0187] The training set was directly fitted using the constructed multi-period and trend term models, and the results are as follows: Figure 4 As shown, the first 86,400 data points fitted by the established multi-period and trend term model perfectly match the original data in the training set, proving the correctness of the model. Meanwhile, comparing the last 86,400 data points predicted by the established multi-period and trend term model with the original data in the test set, it is evident that the prediction deviation increases significantly over time, proving that the Rowland time delay signal has undergone significant drift over time. Statistical analysis of the output of the multi-period and trend term model shows that 70.38% of the points in the test set have a prediction error <50ns, indicating a large error and requiring correction of the model's output.

[0188] Therefore, the residual signal between the observed values ​​of the initial Rowland time-delay signal and the output of the multi-period and trend term models is:

[0189]

[0190] Wavelet decomposition and thresholding denoising are performed on the residual signal. The Symlet4 wavelet is chosen as the wavelet basis function, and the decomposition level is [not specified]. =5, threshold denoising is performed using the SureShrink thresholding method to reconstruct the denoising residual. .

[0191] An adaptive weighting mechanism is used to fuse the outputs of the multi-period and trend term models with the denoised residual signal to obtain the predicted Rowland time delay signal. This includes weight initialization. , =0.1, =1.0, γ=0.05, weights are updated every 100 points, based on the ratio of recent error to historical error:

[0192]

[0193] Further, in order to improve the prediction accuracy of the model, in the prediction process, the trend term is refitted every 1 hour (3600 seconds) through the time window, and the data in the window is the latest 3600 points. The new trend term is linearly mixed with the old trend term, and the transition coefficient β=0.1, and the final prediction result is as shown in Figure 5 As shown, the statistics use the combined prediction results based on the multi-period term modeling and wavelet analysis, and the proportion of points with a prediction error of less than 50 ns in the statistical test set is 91.39%.

[0194] The predicted Loran time delay signal obtained by fusing the initial Loran time delay signal, the output of the multi-period term and trend term model, and the output of the multi-period term and trend term model and the denoised residual signal is compared, as shown in Figure 6 As shown in Figures 4-6 It can be seen that the prediction result based on the prediction method of the embodiment has been obviously improved compared with the result of only model prediction, and the proportion of points with a prediction error of less than 50 ns is increased from 70.38% to 91.39%.

[0195] An exemplary embodiment of the present application also provides a Loran time delay signal prediction system, as shown in Figure 7 The system comprises:

[0196] The acquisition module 71 is configured to acquire an initial Loran time delay signal.

[0197] The extraction module 72 is configured to extract a periodic term of the initial Loran time delay signal and input the periodic term into a constructed multi-period term and trend term model; the multi-period term and trend term model is as follows:

[0198]

[0199] Wherein, ωi is the frequency of the i th periodic term, t is the current time, and N is the number of periodic terms, is the sine amplitude of the i th periodic term, is the cosine amplitude of the i th periodic term, is the trend slope, and is a constant term. The residual calculation module 73 is configured to calculate a residual signal between an observation value of the initial Loran time delay signal and an output of the multi-period term and trend term model; the residual signal is as follows:

[0200]

[0201]

[0202] Wherein, e is the residual signal, is the observation value of the initial Loran time delay signal, is the output of the multi-period term and trend term model. ​​

[0203] Decomposition and denoising module 74 is used to perform wavelet decomposition and threshold denoising on the residual signal; the denoised residual signal is:

[0204]

[0205] in, The signal is the denoised residual signal, and J is the number of wavelet decomposition levels. The approximation coefficients for the Jth layer are... denoised as the detail coefficients of the j-th layer.

[0206] The fusion module 75 is used to fuse the outputs of the multi-period and trend term models and the denoised residual signals through an adaptive weighting mechanism to obtain the predicted Rowland time delay signal; the adaptive weighting mechanism is as follows:

[0207]

[0208] in, To predict the Rowland delay signal, ∈[0,1] represents the adaptive weights.

[0209] In one embodiment, the system further includes:

[0210] The mean-reduction module is used to perform mean-reduction processing on the initial Rowland delay signal; the initial Rowland delay signal after mean-reduction processing... for:

[0211]

[0212] The initial Rowland time delay signal is: M is the number of sampling points. The sampling interval is... For the k-th sampling point, For the observation value corresponding to the kth sampling point, The mean of the initial Rowland time delay signal for all sampling points.

[0213] In one embodiment, the wavelet denoising module is also used for:

[0214] Symlet4 wavelets are used as the basis functions for wavelet decomposition;

[0215] In one embodiment, the number of decomposition levels J is:

[0216] ;

[0217] Where M is the number of sampling points of the initial Rowland time delay signal.

[0218] In one embodiment, the wavelet denoising module is also used for:

[0219] The detail coefficients are denoised by threshold method Threshold is:

[0220]

[0221] Wherein, is the standard deviation of the jth layer noise, , is the median absolute deviation of . .

[0222] The detail coefficients are soft-thresholded; the formula is:

[0223] Wherein,

[0224] is the jth layer detail coefficient after soft-thresholding, and k is the coefficient index. In one embodiment, the adaptive weight

[0225] is:

[0226]

[0227] Wherein, is the maximum weight, γ is the adaptive rate, is the average absolute error of the last W sampling points, is the historical average error.

[0228] In one embodiment, the multi-period term and the trend term model update the trend term through a time window; the update formula of the trend term is:

[0229]

[0230] Wherein, is the updated trend slope, is the updated constant term, is the time window, is the kth sampling point of the initial Loran time delay signal.

[0231] ​For system embodiments, as they substantially correspond to the method embodiments, reference is made to the relevant parts of the method embodiments. The system embodiments described above are merely illustrative, wherein the units described as separate components can or can not be physically separate, and the components of the units can or can not be physical units, i.e., can be located in one place, or can be distributed on multiple network units. Some or all of the modules can be selected according to actual needs to achieve the purposes of the present disclosure.

[0232] An exemplary embodiment of the present application also provides an electronic device, referring to Figure 8 The electronic device 800 can have great differences in configuration or performance, and can include one or more central processing units (CPUs) 810 (the processor 810 can include, but is not limited to, a microprocessor MCU or a programmable logic device FPGA processing device), a memory 830 for storing data, one or more storage media 820 (such as one or more mass storage devices) for storing application programs 823 or data 822. Among them, the memory 830 and the storage medium 820 can be temporary storage or persistent storage. The storage medium 820 stores at least one instruction, at least one program, a code set or an instruction set, which is loaded and executed by the processor to implement the above-mentioned prediction method of the Loran time delay signal.

[0233] Further, the central processing unit 810 can be configured to communicate with the storage medium 820 and execute a series of instruction operations in the storage medium 820 on the electronic device 800. The electronic device 800 can also include one or more power supplies 860, one or more wired or wireless network interfaces 850, one or more input / output interfaces 840, and / or one or more operating systems 821, such as Windows ServerTM, Mac OSXTM, UnixTM, LinuxTM, FreeBSDTM, etc.

[0234] The input / output interface 840 can be used to receive or send data via a network. The above-mentioned network examples can include a wireless network provided by a communication provider of the electronic device 800. In one example, the input / output interface 840 includes a network adapter (NIC) that can be connected to other network devices through a base station to communicate with the Internet. In one example, the input / output interface 840 can be a radio frequency (RF) module for communicating with the Internet in a wireless manner.

[0235] Those skilled in the art can understand that Figure 8 The structure shown is only schematic, and does not limit the structure of the electronic device described above. For example, the electronic device 800 can also include more or less components than those shown, or have a different configuration of components than shown. Figure 8 The structure shown is only schematic, and does not limit the structure of the electronic device described above. For example, the electronic device 800 can also include more or less components than those shown, or have a different configuration of components than shown. Figure 8 The structure shown is only schematic, and does not limit the structure of the electronic device described above. For example, the electronic device 800 can also include more or less components than those shown, or have a different configuration of components than shown.

[0236] The embodiment of the present application also provides a computer readable storage medium, which can be arranged in a server to store at least one instruction, at least one program, a code set or an instruction set related to a prediction method of a Loran time delay signal, the at least one instruction, the at least one program, the code set or the instruction set being loaded and executed by the processor to realize the prediction method of the Loran time delay signal.

[0237] Optionally, in the embodiment, the storage medium can be located in at least one network server of a computer network. Optionally, in the embodiment, the storage medium can include but is not limited to a U disk, a read-only memory (ROM), a random access memory (RAM), a mobile hard disk, a magnetic disk or an optical disk and various storage program codes.

[0238] The embodiment of the present application also provides a computer program product, which includes a computer program, and the computer program is executed by the processor to realize the prediction method of the Loran time delay signal.

[0239] The program code of the computer program product for executing the present disclosure can be written in any combination of one or more programming languages, and can be executed completely on the user device, partially on the user device, as an independent software package, partially on the user device and partially on a remote device, or completely on a remote device.

[0240] It can be known from common technical knowledge that the present application can be realized by other embodiments without departing from the spirit or essential characteristics thereof. Therefore, the above disclosed embodiments are only examples, and are not the only ones. All changes within the scope of the present application or within the scope equivalent to the present application are included in the present application.

[0241] Those skilled in the art will appreciate that embodiments of the application can be devised for a method, a system, or a computer program product. Accordingly, the present application can be embodied in the form of an entirely hardware embodiment, an entirely software embodiment, or an embodiment combining software and hardware aspects. Furthermore, the present application can take the form of a computer program product on one or more computer-usable storage media (including, but not limited to, disk storage, CD-ROMs, optical storage devices, etc.) embodying computer readable program code.

[0242] The present application is described in reference to the flowchart and / or block diagrams of the method, apparatus (system) and computer program product according to embodiments of the application. It will be understood that each block of the flowchart and / or block diagrams, and combinations of blocks in the flowchart and / or block diagrams, can be implemented by computer program instructions. These computer program instructions can be provided to a processor of a general purpose computer, special purpose computer, embedded processing device or other programmable data processing apparatus to produce a machine, such that the instructions, which execute via the processor of the computer or other programmable data processing apparatus, create means for implementing the functions specified in the flowchart and / or block diagram block or blocks. Figure One one or more functions specified in the flowchart and / or block diagram block or blocks. Figure One one or more functions specified in the flowchart and / or block diagram block or blocks.

[0243] These computer program instructions can also be stored in a computer- readable memory that can direct a computer or other programmable data processing apparatus to function in a particular manner, such that the instructions stored in the computer-readable memory produce an article of manufacture including instructions which implement the function specified in the flowchart and / or block diagram block or blocks. Figure One one or more functions specified in the flowchart and / or block diagram block or blocks. Figure One one or more functions specified in the flowchart and / or block diagram block or blocks.

[0244] These computer program instructions can also be loaded onto a computer or other programmable data processing apparatus to cause a series of operational steps to be performed on the computer or other programmable apparatus to produce a computer-implemented process such that the instructions which execute on the computer or other programmable apparatus provide steps for implementing the functions specified in the flowchart and / or block diagram block or blocks. Figure One one or more functions specified in the flowchart and / or block diagram block or blocks. Figure One one or more functions specified in the flowchart and / or block diagram block or blocks.

[0245] Finally, it should be noted that the above-mentioned embodiments are only intended to illustrate the technical solutions of the present application, and are not intended to limit the present application. Although the present application has been described in detail with reference to the above-mentioned embodiments, those skilled in the art should understand that the specific embodiments of the present application can be modified or replaced by equivalents without departing from the spirit and scope of the present application, and any modifications or equivalents within the spirit and scope of the present application should be covered by the claims of the present application.

Claims

1. A method of predicting a Loran time delay signal, characterized by, The method comprises: acquiring an initial Loran time delay signal; extracting the periodic term of the initial Loran delay signal, inputting into the constructed multi-periodic term and trend term model; the multi-periodic term and trend term model is: wherein, is the frequency of the ith periodic term, t is the current time, N is the number of periodic terms, is the sine amplitude of the ith periodic term, is the cosine amplitude of the ith periodic term, is the trend slope, and is the constant term; computing a residual signal between the observations of the initial Loran time delay signal and the output of the multi-periodic term and trend term model; the residual signal is: wherein, is an observation of an initial Loran time delay signal, is an output of the multi-periodic term and trend term model; Wavelet decomposition and threshold denoising are performed on the residual signal; the residual signal after denoising is: wherein J is the number of decomposition levels of wavelet decomposition, is the jth approximation coefficient of the Jth level, is the jth detail coefficient of the Jth level after soft threshold processing. The output of the multi-period term and the trend term model and the denoised residual signal are fused through an adaptive weight mechanism to obtain a predicted Loran delay signal; the predicted Loran delay signal is is: wherein, ∈ [0,1] is an adaptive weight.

2. The method of predicting a Loran time delay signal as claimed in claim 1, wherein, Before inputting the period term of the initial Loran time delay signal to the constructed multi-period term and trend term model, the method further comprises: The initial Loran time delay signal is subjected to a de-meaning process; the de-meaned initial Loran time delay signal is is: wherein the initial Loran time delay signal is M is the number of sampling points of the initial Loran time delay signal, is a sampling interval, is the kth sampling point, is an observation value corresponding to the kth sampling point, is the mean value of the initial Loran time delay signal of all sampling points.

3. The method for predicting Rowland time-delay signals as described in claim 1, characterized in that, The wavelet decomposition and threshold denoising of the residual signal comprises: Symlet4 wavelet is used as the base function of wavelet decomposition; And / or, the decomposition layer number J is: Wherein, M is the number of sampling points of the initial Loran time delay signal.

4. The method for predicting Rowland time-delay signals as described in claim 1, characterized in that, The wavelet decomposition and threshold denoising of the residual signal comprises: The threshold method is used to denoise the jth layer of detail coefficients is:​ wherein is the standard deviation of the noise of the jth layer, , is the median absolute deviation of ;​ for j = 1 to N Soft thresholding; formula: Wherein, k is the coefficient index.

5. The method for predicting Rowland time-delay signals as described in claim 1, characterized in that, The adaptive weights are: where, is the maximum weight, γ is the adaptation rate, is the average absolute error of the last W samples, is the historical average error.

6. The method for predicting Rowland time-delay signals as described in claim 1, characterized in that, The multi-period term and trend term model updates the trend term through a time window; the update formula of the trend term is: wherein, is a periodic term, is an updated trend slope, is an updated constant term, is a time window, is the kth sample point of the initial Loran time delay signal.

7. A system for predicting Loran time delay signals, characterized by The system comprises: an acquisition module configured to acquire an initial Loran time delay signal; The extraction module extracts the periodic term of the initial Loran time delay signal and inputs the periodic term into a constructed multi-periodic term and trend term model; the multi-periodic term and trend term model is: wherein, is the frequency of the ith periodic term, t is the current time, N is the number of periodic terms, is the sine amplitude of the ith periodic term, is the cosine amplitude of the ith periodic term, is the trend slope, and is the constant term; a residual calculation module configured to calculate a residual signal between an observation of the initial Loran time delay signal and an output of the multi-periodic term and trend term model; the residual signal is: wherein, is an observation of an initial Loran time delay signal, is an output of the multi-periodic term and trend term model; The wavelet denoising module is configured to perform wavelet decomposition and threshold denoising on the residual signal; and the denoised residual signal is: wherein J is the number of decomposition levels of wavelet decomposition, is the jth approximation coefficient of the Jth level, is the jth detail coefficient of the Jth level after soft threshold processing. a fusion module configured to fuse the multi-periodic term and the output of the trend term model and the denoised residual signal by an adaptive weight mechanism to obtain a predicted Loran delay signal; and is: wherein, ∈ [0,1] is an adaptive weight.

8. An electronic device, comprising: The electronic device comprises a processor and a memory, and the memory stores at least one instruction, at least one program, a code set or an instruction set, and the processor loads and executes the at least one instruction, the at least one program, the code set or the instruction set to implement the Loran time delay signal prediction method according to any one of claims 1-6.

9. A computer-readable storage medium, characterized in that, The computer readable storage medium stores at least one instruction or at least one program, and the processor loads and executes the at least one instruction or the at least one program to implement the Loran time delay signal prediction method according to any one of claims 1-6.

10. A computer program product comprising a computer program, characterized in that, The computer program is executed by the processor to implement the Loran time delay signal prediction method according to any one of claims 1-6.

Citation Information

Patent Citations

  • Timing method and device based on Rowland signal tracking and terminal equipment

    CN117608181A

  • Trend Kriging-based ASF estimation method for improving positioning and navigation precision of Rowland system

    CN119805362A