Model Generation, Significant Wave Height Inversion Method, Apparatus, Device, and Storage Medium

Abstract

The embodiments of the present application provide a method, device, equipment, and storage medium for model generation and significant wave height inversion. The model generation method includes: obtaining a first measured GNSS-R normalized one-dimensional delay waveform and obtaining a first theoretical GNSS-R normalized one-dimensional delay waveform; obtaining a first delay range difference corresponding to the same amplitude range of the first measured GNSS-R normalized one-dimensional delay waveform and the first theoretical GNSS-R normalized one-dimensional delay waveform; performing incident angle correction on the first delay range difference to obtain a first delay difference observation; obtaining a first significant wave height that matches the first delay difference observation; and training a significant wave height inversion model using the first delay difference observation and the first significant wave height. The embodiments of the present application improve the accuracy of the significant wave height inversion model, and thus can improve the accuracy of the significant wave height inverted by the significant wave height inversion model.

Description

Technical Field

[0001] This application relates to the field of reflective remote sensing technology, and particularly to a method, apparatus, device, and storage medium for model generation and significant wave height inversion. Background Art

[0002] The significant wave height is an important ocean wave parameter, representing the statistical value of the actual wave height. It is defined as the average value of the first 1 / 3 of the wave heights after arranging the wave heights in descending order, and is of great significance for ocean disaster warning and emergency management.

[0003] Currently, the remote sensing means for quantitatively measuring the significant wave height include radar altimeters, ocean wave spectrometers, and SAR (Synthetic Aperture Radar). However, currently, there is less remote sensing observation data for the significant wave height, mainly because radar altimeters and ocean wave spectrometers can only observe the sub-satellite point of the satellite, and the observation is only strip-shaped, lacking a wide-swath measurement method. And due to high power consumption, SAR cannot continuously observe, and has a narrow swath, resulting in low accuracy of the significant wave height. Summary of the Invention

[0004] In view of the above problems, embodiments of this application propose a method, apparatus, device, and storage medium for model generation and significant wave height inversion to improve the accuracy of the significant wave height.

[0005] According to one aspect of the embodiments of this application, a method for generating a significant wave height inversion model is provided, the method including:

[0006] Obtaining a first measured GNSS-R normalized one-dimensional time-delay waveform, and obtaining a first theoretical GNSS-R normalized one-dimensional time-delay waveform;

[0007] Obtaining a first time-delay range difference corresponding to the same amplitude range of the first measured GNSS-R normalized one-dimensional time-delay waveform and the first theoretical GNSS-R normalized one-dimensional time-delay waveform;

[0008] Performing incident angle correction on the first time-delay range difference to obtain a first time-delay difference observation;

[0009] Obtaining a first significant wave height that matches the first time-delay difference observation;

[0010] Training an effective wave height inversion model by using the first time-delay difference observation and the first significant wave height.

[0011] Optionally, obtaining the first measured GNSS-R normalized one-dimensional delay waveform includes: obtaining the first measured DDM waveform; performing denoising processing on the first measured DDM waveform to obtain a first target measured DDM waveform; extracting the value corresponding to a Doppler frequency of 0 in the first target measured DDM waveform as a first target value, and dividing the first target value by the peak value in the first target value to obtain the first measured GNSS-R normalized one-dimensional delay waveform.

[0012] Optionally, the performing denoising processing on the first measured DDM waveform to obtain a first target measured DDM waveform includes: calculating a first average value of the values corresponding to the delays less than or equal to a preset delay threshold in the first measured DDM waveform, and using the first average value as the noise floor of the first measured DDM waveform; subtracting the noise floor of the first measured DDM waveform from the first measured DDM waveform to obtain the first target measured DDM waveform.

[0013] Optionally, obtaining the first theoretical GNSS-R normalized one-dimensional delay waveform includes: obtaining the first theoretical DDM waveform; performing denoising processing on the first theoretical DDM waveform to obtain a first target theoretical DDM waveform; extracting the value corresponding to a Doppler frequency of 0 in the first target theoretical DDM waveform as a second target value, and dividing the second target value by the peak value in the second target value to obtain the first theoretical GNSS-R normalized one-dimensional delay waveform.

[0014] Optionally, the performing denoising processing on the first theoretical DDM waveform to obtain a first target theoretical DDM waveform includes: calculating a second average value of the values corresponding to the delays less than or equal to a preset delay threshold in the first theoretical DDM waveform, and using the second average value as the noise floor of the first theoretical DDM waveform; subtracting the noise floor of the first theoretical DDM waveform from the first theoretical DDM waveform to obtain the first target theoretical DDM waveform.

[0015] Optionally, the obtaining of the first time delay range difference corresponding to the same amplitude range between the first measured GNSS-R normalized one-dimensional time delay waveform and the first theoretical GNSS-R normalized one-dimensional time delay waveform includes: obtaining the first theoretical specular reflection point of the first theoretical GNSS-R normalized one-dimensional time delay waveform, and obtaining the first amplitude range corresponding to the first time delay range centered on the first theoretical specular reflection point in the first theoretical GNSS-R normalized one-dimensional time delay waveform; obtaining the first measured specular reflection point of the first measured GNSS-R normalized one-dimensional time delay waveform, and obtaining the second time delay range centered on the first measured specular reflection point corresponding to the first measured GNSS-R normalized one-dimensional time delay waveform under the first amplitude range; calculating the difference between the second time delay range and the first time delay range to obtain the first time delay range difference.

[0016] Optionally, the obtaining of the first amplitude range corresponding to the first time delay range centered on the first theoretical specular reflection point in the first theoretical GNSS-R normalized one-dimensional time delay waveform includes: performing linear fitting on the waveform of the first time delay range centered on the first theoretical specular reflection point in the first theoretical GNSS-R normalized one-dimensional time delay waveform to obtain a first theoretical fitting waveform; obtaining the amplitude range corresponding to the first time delay range in the first theoretical fitting waveform as the first amplitude range.

[0017] Optionally, the obtaining of the first measured specular reflection point of the first measured GNSS-R normalized one-dimensional time delay waveform includes:

[0018] Selecting a preset time delay offset one by one;

[0019] Calculating the following cost function respectively according to each time delay offset:

[0020]

[0021] where J(k) represents the cost function, represents the normalized value corresponding to the time delay τ i+k in the first measured GNSS-R normalized one-dimensional time delay waveform, represents the normalized value corresponding to the time delay τ i in the first theoretical GNSS-R normalized one-dimensional time delay waveform, k represents the time delay offset, N1 represents the subscript of the lower limit of the time delay of the third time delay range, and N2 represents the subscript of the upper limit of the time delay of the third time delay range;

[0022] Select the target delay offset corresponding to the minimum cost function, and use the delay corresponding to the first theoretical specular reflection point offset by the target delay offset as the measured specular reflection point of the first measured GNSS-R normalized one-dimensional delay waveform.

[0023] Optionally, the obtaining of the second delay range centered on the first measured specular reflection point corresponding to the first measured GNSS-R normalized one-dimensional delay waveform in the first amplitude range includes: linearly fitting the waveform of the first delay range centered on the first measured specular reflection point in the first measured GNSS-R normalized one-dimensional delay waveform to obtain a first measured fitting waveform; calculating the delay range corresponding to the first amplitude range in the first measured fitting waveform as the second delay range.

[0024] Optionally, the correcting the first delay range difference for the incident angle to obtain a first delay difference observation includes: determining the ratio obtained by dividing the first delay range difference by twice the cosine value of the first incident angle as the first delay difference observation.

[0025] Optionally, the obtaining of the first significant wave height matching the first delay difference observation includes: matching the first delay difference observation with the significant wave height measured by the buoy according to a set matching rule to obtain a first significant wave height matching the first delay difference observation; the matching rule includes: the spatial distance between the specular reflection point position corresponding to the GNSS-R measurement and the buoy position corresponding to the buoy measurement is less than a preset distance, and the time interval between the observation time corresponding to the GNSS-R measurement and the observation time corresponding to the buoy measurement is less than a preset duration.

[0026] Optionally, the training of the significant wave height inversion model using the first delay difference observation and the first significant wave height includes:

[0027] Performing linear fitting training using the first delay difference observation and the first significant wave height to obtain the following significant wave height inversion model:

[0028] H s = aΔh + b

[0029] where H S represents the significant wave height, Δh represents the delay difference observation, a represents the first coefficient, and b represents the second coefficient.

[0030] According to another aspect of the embodiments of the present application, there is provided a significant wave height inversion method, the method including:

[0031] Obtain the second measured GNSS-R normalized one-dimensional time-delay waveform and obtain the second theoretical GNSS-R normalized one-dimensional time-delay waveform;

[0032] Obtain the second time-delay range difference corresponding to the same amplitude range of the second measured GNSS-R normalized one-dimensional time-delay waveform and the second theoretical GNSS-R normalized one-dimensional time-delay waveform;

[0033] Perform incidence angle correction on the second time-delay range difference to obtain the second time-delay difference observation;

[0034] Use the pre-trained significant wave height inversion model to calculate the significant wave height corresponding to the second time-delay difference observation; the significant wave height inversion model is obtained by the significant wave height inversion model generation method described in any one of the above.

[0035] According to another aspect of the embodiments of the present application, there is provided a significant wave height inversion model generation device, the device includes:

[0036] The first acquisition module is used to acquire the first measured GNSS-R normalized one-dimensional time-delay waveform and acquire the first theoretical GNSS-R normalized one-dimensional time-delay waveform;

[0037] The second acquisition module is used to acquire the first time-delay range difference corresponding to the same amplitude range of the first measured GNSS-R normalized one-dimensional time-delay waveform and the first theoretical GNSS-R normalized one-dimensional time-delay waveform;

[0038] The first correction module is used to perform incidence angle correction on the first time-delay range difference to obtain the first time-delay difference observation;

[0039] The matching module is used to acquire the first significant wave height that matches the first time-delay difference observation;

[0040] The training module is used to train the significant wave height inversion model by using the first time-delay difference observation and the first significant wave height.

[0041] Optionally, the first acquisition module includes: a first waveform acquisition sub-module for acquiring the first measured DDM waveform; a first denoising sub-module for denoising the first measured DDM waveform to obtain the first target measured DDM waveform; a first normalization sub-module for extracting the value corresponding to the Doppler frequency of 0 in the first target measured DDM waveform as the first target value, and dividing the first target value by the peak value in the first target value to obtain the first measured GNSS-R normalized one-dimensional time-delay waveform.

[0042] Optionally, the first denoising module is specifically configured to calculate a first average value of the values corresponding to the time delays less than or equal to a preset time delay threshold in the first measured DDM waveform, and use the first average value as the noise floor of the first measured DDM waveform; subtract the noise floor of the first measured DDM waveform from the first measured DDM waveform to obtain the first target measured DDM waveform.

[0043] Optionally, the first obtaining module includes: a second waveform obtaining sub-module, configured to obtain a first theoretical DDM waveform; a second denoising sub-module, configured to perform denoising processing on the first theoretical DDM waveform to obtain a first target theoretical DDM waveform; a second normalization sub-module, configured to extract a value corresponding to a Doppler frequency of 0 in the first target theoretical DDM waveform as a second target value, and divide the second target value by a peak value in the second target value to obtain the first theoretical GNSS-R normalized one-dimensional time delay waveform.

[0044] Optionally, the second denoising sub-module is specifically configured to calculate a second average value of the values corresponding to the time delays less than or equal to a preset time delay threshold in the first theoretical DDM waveform, and use the second average value as the noise floor of the first theoretical DDM waveform; subtract the noise floor of the first theoretical DDM waveform from the first theoretical DDM waveform to obtain the first target theoretical DDM waveform.

[0045] Optionally, the second obtaining module includes: a first range obtaining sub-module, configured to obtain a first theoretical specular reflection point of the first theoretical GNSS-R normalized one-dimensional time delay waveform, and obtain a first amplitude range corresponding to a first time delay range centered on the first theoretical specular reflection point in the first theoretical GNSS-R normalized one-dimensional time delay waveform; a second range obtaining sub-module, configured to obtain a first measured specular reflection point of the first measured GNSS-R normalized one-dimensional time delay waveform, and obtain a second time delay range centered on the first measured specular reflection point corresponding to the first measured GNSS-R normalized one-dimensional time delay waveform under the first amplitude range; a calculation sub-module, configured to calculate a difference between the second time delay range and the first time delay range to obtain the first time delay range difference.

[0046] Optionally, the first range obtaining sub-module includes: a first fitting unit, configured to perform linear fitting on the waveform of a first time delay range centered on the first theoretical specular reflection point in the first theoretical GNSS-R normalized one-dimensional time delay waveform to obtain a first theoretical fitting waveform; obtain an amplitude range corresponding to the first time delay range in the first theoretical fitting waveform as the first amplitude range.

[0047] Optionally, the second range obtaining sub-module includes a reflection point obtaining unit, and the reflection point obtaining unit is configured to:

[0048] Select the preset time delay offsets one by one;

[0049] Calculate the following cost function according to each of the time delay offsets respectively:

[0050]

[0051] where J(k) represents the cost function, represents the normalized value corresponding to the time delay τ in the first measured GNSS-R normalized one-dimensional time delay waveform i+k corresponding thereto, represents the normalized value corresponding to the time delay τ in the first theoretical GNSS-R normalized one-dimensional time delay waveform i corresponding thereto, k represents the time delay offset, N1 represents the subscript of the lower limit of the time delay in the third time delay range, and N2 represents the subscript of the upper limit of the time delay in the third time delay range;

[0052] Select the target time delay offset corresponding to the minimum of the cost function, and use the time delay corresponding to the first theoretical specular reflection point offset by the target time delay offset as the measured specular reflection point of the first measured GNSS-R normalized one-dimensional time delay waveform.

[0053] Optionally, the first range acquisition sub-module includes: a second fitting unit, configured to perform linear fitting on the waveform of the first time delay range centered on the first measured specular reflection point in the first measured GNSS-R normalized one-dimensional time delay waveform to obtain a first measured fitting waveform; calculate the time delay range corresponding to the first amplitude range in the first measured fitting waveform as the second time delay range.

[0054] Optionally, the first correction module is specifically configured to determine the ratio obtained by dividing the first time delay range difference by twice the cosine value of the first incident angle as the first time delay difference observable.

[0055] Optionally, the matching module is specifically configured to match the first time delay difference observable with the significant wave height measured by the buoy according to a set matching rule to obtain a first significant wave height matching the first time delay difference observable; the matching rule includes: the spatial distance between the specular reflection point position corresponding to the GNSS-R measurement and the buoy position corresponding to the buoy measurement is less than a preset distance, and the time interval between the observation time corresponding to the GNSS-R measurement and the observation time corresponding to the buoy measurement is less than a preset duration.

[0056] Optionally, the training module is specifically configured to perform linear fitting training using the first time delay difference observable and the first significant wave height to obtain the following significant wave height inversion model:

[0057] H s = aΔh + b

[0058] Wherein, H S represents the significant wave height, Δh represents the observed value of the time delay difference, a represents the first coefficient, and b represents the second coefficient.

[0059] According to another aspect of the embodiments of the present application, there is provided an apparatus for inverting the significant wave height, the apparatus including:

[0060] A third acquisition module, configured to acquire the second measured GNSS-R normalized one-dimensional time delay waveform and acquire the second theoretical GNSS-R normalized one-dimensional time delay waveform;

[0061] A fourth acquisition module, configured to acquire the second time delay range difference corresponding to the same amplitude range of the second measured GNSS-R normalized one-dimensional time delay waveform and the second theoretical GNSS-R normalized one-dimensional time delay waveform;

[0062] A second correction module, configured to perform incident angle correction on the second time delay range difference to obtain a second observed value of the time delay difference;

[0063] A calculation module, configured to calculate the significant wave height corresponding to the second observed value of the time delay difference by using a pre-trained significant wave height inversion model; the significant wave height inversion model is obtained by the significant wave height inversion model generation method described in any one of the above.

[0064] According to another aspect of the embodiments of the present application, there is provided an electronic device, the electronic device including a processor and a computer-readable storage medium, and a computer program is stored on the computer-readable storage medium; when the computer program is executed by the processor, the processor is caused to execute the significant wave height inversion model generation method described in any one of the above, or execute the significant wave height inversion method described in any one of the above.

[0065] According to another aspect of the embodiments of the present application, there is provided a computer-readable storage medium, and a computer program is stored on the computer-readable storage medium. When the computer program is executed by a processor, the processor is caused to execute the significant wave height inversion model generation method described in any one of the above, or execute the significant wave height inversion method described in any one of the above.

[0066] In the embodiments of the present application, an effective wave height inversion based on GNSS-R (Global Navigation Satellite System Reflectometry) technology is proposed. The original waveform is normalized to obtain the GNSS-R normalized one-dimensional time-delay waveform. The time-delay information rather than the power amplitude information of the GNSS-R normalized one-dimensional time-delay waveform is used for comparative analysis to obtain the first time-delay difference observation. Then, based on the first time-delay difference observation and the matched first effective wave height, an effective wave height inversion model is trained. Using time-delay information for analysis, compared with using power amplitude information for analysis, does not require signal power calibration, is more direct and has higher accuracy, and does not have problems such as decreased sensitivity in high sea states and reduced accuracy when swells are dominant. Moreover, when calculating the first time-delay difference observation, an incident angle correction is added, further improving the accuracy of the first time-delay difference observation, thereby improving the accuracy of the effective wave height inversion model, and further being able to improve the accuracy of the effective wave height inverted by the effective wave height inversion model. BRIEF DESCRIPTION OF THE DRAWINGS

[0067] In order to more clearly illustrate the technical solutions of the embodiments of the present application, the drawings required for the description of the embodiments of the present application will be briefly introduced below. Obviously, the drawings in the following description are only some drawings of the present application. For those of ordinary skill in the art, other drawings can be obtained based on these drawings without creative efforts.

[0068] Figure 1 is a schematic diagram of the principle of an altimeter and GNSS-R in the embodiments of the present application;

[0069] Figure 2 is a flowchart of the steps of a method for generating an effective wave height inversion model in the embodiments of the present application;

[0070] Figure 3 is a schematic diagram of a GNSS-R normalized one-dimensional time-delay waveform in the embodiments of the present application;

[0071] Figure 4 is a schematic diagram of the linear fitting of a GNSS-R normalized one-dimensional time-delay waveform in the embodiments of the present application;

[0072] Figure 5 is a schematic diagram of a fitted waveform in the embodiments of the present application;

[0073] Figure 6 is a schematic diagram of the change in propagation distance caused by the change in sea surface height in the embodiments of the present application;

[0074] Figure 7 is a flowchart of the steps of an effective wave height inversion method in the embodiments of the present application;

[0075] Figure 8 is a structural block diagram of an effective wave height inversion model generation device according to an embodiment of the present application;

[0076] Figure 9 is a structural block diagram of an effective wave height inversion device according to an embodiment of the present application;

[0077] Figure 10 is a structural block diagram of an electronic device according to an embodiment of the present application;

[0078] Figure 11 is a structural block diagram of a computer-readable storage medium according to an embodiment of the present application. Detailed implementation manners

[0079] To make the objectives, technical solutions, and advantages of the embodiments of the present application clearer, the technical solutions in the embodiments of the present application will be clearly and completely described below with reference to the accompanying drawings in the embodiments of the present application. Apparently, the described embodiments are some, but not all, of the embodiments of the present application. All other embodiments obtained by those of ordinary skill in the art based on the embodiments in the present application without creative efforts shall fall within the protection scope of the present application.

[0080] In the embodiments of the present application, in view of the defects of measuring the effective wave height by a radar altimeter, a sea wave spectrometer, and an SAR method, an effective wave height inversion based on GNSS-R is proposed.

[0081] GNSS-R is a relatively new remote sensing means. This technology utilizes the reflected signals of global navigation satellites (such as the Beidou system, GPS (Global Positioning System), Galileo, etc.) on the Earth's surface to detect physical parameters of the Earth's surface, such as sea surface wind speed, sea surface height, soil moisture, sea ice parameters, etc. Spaceborne GNSS-R uses low-Earth orbit satellites and can receive signals from more than 100 navigation satellites, having the advantages of sufficient signal sources and global coverage. Since the navigation satellites transmit L-band signals, which have strong penetration power and are not affected by clouds, rain, and precipitation, all-weather observations can be provided. In addition, since GNSS-R uses the navigation satellite signals as an opportunity signal and the payload does not need to actively transmit signals, it has the advantages of low power consumption, small volume, and light weight, and is suitable for being carried by small satellites to form a network observation to provide global high spatio-temporal resolution observation data.

[0082] Figure 1 is a schematic diagram of the principle of an altimeter and GNSS-R according to an embodiment of the present application. As Figure 1 shown, compared with the altimeter, GNSS-R can simultaneously utilize multiple specular reflection points for observation, and then obtain observation data of multiple strips within a swath with a width of 1500 km to 2500 km.

[0083] Spaceborne GNSS-R is a potential new significant wave height measurement technology. Since GNSS-R is very suitable for constellation networking detection, it can complement existing observation means and provide global significant wave height observation data with high spatio-temporal resolution.

[0084] In the embodiments of the present application, considering that the spaceborne GNSS-R significant wave height inversion method is based on observables such as the normalized radar cross section and the leading edge slope of the unnormalized waveform, that is, an indirect inversion method based on sea surface roughness. Since the significant wave height is affected by both wind waves and swells, and the sea surface roughness of GNSS-R mainly reflects wind wave information, this method has certain limitations, such as a decrease in sensitivity in high sea states and a reduction in accuracy when swells dominate. Therefore, in the embodiments of the present application, the time delay information of GNSS-R, that is, the range observation value, is used to directly measure the significant wave height, and higher inversion accuracy can be obtained theoretically.

[0085] The idea of the embodiments of the present application is to first normalize the original waveform to calculate the GNSS-R normalized one-dimensional time delay waveform, and compare the theoretical GNSS-R normalized one-dimensional time delay waveform and the measured GNSS-R normalized one-dimensional time delay waveform near the specular reflection point. Due to the influence of sea waves, the slope of the measured GNSS-R normalized one-dimensional time delay waveform is lower than that of the theoretical GNSS-R normalized one-dimensional time delay waveform, and the time delay range corresponding to the same amplitude is larger. Calculate the time delay range difference between the measured GNSS-R normalized one-dimensional time delay waveform and the theoretical GNSS-R normalized one-dimensional time delay waveform corresponding to the same amplitude, and then correct it with the incident angle to obtain the time delay difference observable. Finally, match the time delay difference observable with the significant wave height to train the significant wave height inversion model.

[0086] Refer to Figure 2 , which shows the flowchart of the steps of a significant wave height inversion model generation method according to the embodiments of the present application.

[0087] As Figure 2 shown, the significant wave height inversion model generation method may include the following steps:

[0088] Step 201, obtain the first measured GNSS-R normalized one-dimensional time delay waveform and obtain the first theoretical GNSS-R normalized one-dimensional time delay waveform.

[0089] In the embodiments of the present application, the first measured GNSS-R normalized one-dimensional delay waveform refers to the measured GNSS-R normalized one-dimensional delay waveform used in the generation process of the significant wave height inversion model, and the first theoretical GNSS-R normalized one-dimensional delay waveform refers to the theoretical GNSS-R normalized one-dimensional delay waveform used in the generation process of the significant wave height inversion model. Therefore, the first measured GNSS-R normalized one-dimensional delay waveform and the first theoretical GNSS-R normalized one-dimensional delay waveform can be referred to as training sample data.

[0090] In an alternative embodiment, the first measured GNSS-R normalized one-dimensional delay waveform can be extracted from the first measured DDM (Delay-Doppler map) waveform.

[0091] The first measured DDM waveform refers to the measured DDM waveform used in the generation process of the significant wave height inversion model.

[0092] The DDM waveform is a basic observable of GNSS-R and is a two-dimensional power waveform in the dimensions of delay (unit: chip, specifically the GNSS signal chip) and Doppler frequency (unit: kHz, etc.). The horizontal axis in the DDM waveform represents the delay, and the vertical axis represents the Doppler frequency. Each combination of delay and Doppler frequency corresponds to a value, which can represent the intensity or power of the reflected signal and can be represented by the depth of color, grayscale value, or specific digital quantity.

[0093] Exemplarily, the process of obtaining the first measured GNSS-R normalized one-dimensional delay waveform may include the following steps A1 to A3:

[0094] Step A1, obtain the first measured DDM waveform.

[0095] In the embodiments of the present application, the DDM waveform is a basic observable of GNSS-R, and the first measured DDM waveform can be directly obtained from the GNSS-R satellite data.

[0096] Step A2, perform denoising processing on the first measured DDM waveform to obtain the first target measured DDM waveform.

[0097] Exemplarily, the process of performing denoising processing on the first measured DDM waveform to obtain the first target measured DDM waveform may include: calculating the first average value of the values corresponding to the delays less than or equal to the preset delay threshold in the first measured DDM waveform, and using the first average value as the noise floor of the first measured DDM waveform; subtracting the noise floor of the first measured DDM waveform from the first measured DDM waveform to obtain the first target measured DDM waveform.

[0098] For example, the first target measured DDM waveform can be obtained using the following formula (1):

[0099] C r C(τ, f) = C(τ, f) - C N (1)

[0100] Wherein, τ represents the time delay, f represents the Doppler frequency, C(τ, f) represents the value corresponding to the time delay τ and the Doppler frequency f in the first measured DDM waveform, and C N represents the noise floor of the first measured DDM waveform, and C r C(τ, f) represents the value corresponding to the time delay τ and the Doppler frequency f in the first target measured DDM waveform.

[0101] For the specific value of the above time delay threshold, it can be set according to actual requirements. For example, the time delay threshold can be set to -1.5 chips, -2 chips, -2.5 chips, etc., and this embodiment does not limit this.

[0102] Step A3, extract the value corresponding to the Doppler frequency of 0 in the first target measured DDM waveform as the first target value, and divide the first target value by the peak value in the first target value to obtain the first measured GNSS-R normalized one-dimensional time delay waveform.

[0103] For example, the first measured GNSS-R normalized one-dimensional time delay waveform can be obtained by using the following formula (2):

[0104]

[0105] Wherein, C r C(τ, 0) represents the value corresponding to the time delay τ when the Doppler frequency is 0 in the first target measured DDM waveform, that is, the first target value, and C r，p represents the peak value in the first target value, represents the normalized value corresponding to the time delay τ in the first measured GNSS-R normalized one-dimensional time delay waveform.

[0106] In the embodiment of the present application, the horizontal axis of the first measured GNSS-R normalized one-dimensional time delay waveform represents the time delay, and the vertical axis represents the normalized value.

[0107] In an alternative embodiment, the first theoretical GNSS-R normalized one-dimensional time delay waveform can be extracted from the first theoretical DDM waveform.

[0108] The first theoretical DDM waveform refers to the theoretical DDM waveform used in the generation process of the significant wave height inversion model.

[0109] Exemplarily, the process of obtaining the first theoretical GNSS-R normalized one-dimensional time delay waveform may include the following steps B1 to B3:

[0110] Step B1: Obtain the first theoretical DDM waveform.

[0111] In the embodiments of the present application, a theoretical simulation model can be used for simulation calculations to obtain the first theoretical DDM waveform. It should be noted that the theoretical simulation model takes into account the influence of wind speed on the waveform and does not consider the influence of significant wave height. Among them, the theoretical simulation module can include, but is not limited to, the Z-V model, statistical models, etc. For the specific process of using the theoretical simulation model for simulation calculations, it can be processed according to actual experience, and this embodiment will not elaborate in detail here.

[0112] Taking the Z-V model as an example, the simulation calculations can be carried out through the following process: Define radar parameters: including transmit frequency, pulse width, pulse repetition frequency, antenna gain, etc. Simulate sea surface scattering: Use the Z-V model or a similar sea surface scattering model to calculate the backscattering coefficient at different incident angles. Calculate the radar echo: Calculate the received echo power according to the radar equation and the sea surface scattering coefficient. Generate DDM: Construct the DDM by simulating the echo signals at different distances and velocities (Doppler frequencies).

[0113] Step B2: Denoise the first theoretical DDM waveform to obtain the first target theoretical DDM waveform.

[0114] Exemplarily, the process of denoising the first theoretical DDM waveform to obtain the first target theoretical DDM waveform can include: calculating the second average value of the values corresponding to the time delays less than or equal to the preset time delay threshold in the first theoretical DDM waveform, and using the second average value as the noise floor of the first theoretical DDM waveform; subtracting the noise floor of the first theoretical DDM waveform from the first theoretical DDM waveform to obtain the first target theoretical DDM waveform.

[0115] In implementation, the first target theoretical DDM waveform can be obtained using the above formula (1), and specific reference can be made to the relevant descriptions above.

[0116] Step B3: Extract the value corresponding to the Doppler frequency of 0 in the first target theoretical DDM waveform as the second target value, and divide the second target value by the peak value in the second target value to obtain the first theoretical GNSS-R normalized one-dimensional time delay waveform.

[0117] In implementation, the first theoretical GNSS-R normalized one-dimensional time delay waveform can be obtained using the above formula (2), and specific reference can be made to the relevant descriptions above.

[0118] In the embodiments of the present application, the horizontal axis of the first theoretical GNSS-R normalized one-dimensional time delay waveform represents time delay, and the vertical axis represents the normalized value.

[0119] Figure 3 This is a schematic diagram of a GNSS-R normalized one-dimensional delay waveform according to an embodiment of the present application. As Figure 3 shown, the horizontal axis of the GNSS-R normalized one-dimensional delay waveform represents the delay, and the vertical axis represents the normalized value.

[0120] For the GNSS-R normalized one-dimensional delay waveform, when the sea surface is completely calm, the waveform front of the GNSS-R normalized one-dimensional delay waveform depends on the autocorrelation function of its own GNSS signal. When there are ocean waves, due to the presence of wave crests and wave troughs in the ocean waves, the reflected echoes of the signal at the wave crests and wave troughs of the ocean waves have different delays. When the signal hits the wave crest, the delay decreases and the waveform advances; when the signal hits the wave trough, the delay increases and the waveform lags. The larger the ocean waves, the larger the amplitude, and the larger the advance and lag amounts of the waveform, so the steeper the slope of the waveform front.

[0121] It should be noted that in the embodiment of the present application, for the delay sampling interval of the first theoretical GNSS-R normalized one-dimensional delay waveform and the first measured GNSS-R normalized one-dimensional delay waveform, it can be set according to actual needs. For example, the delay sampling interval can be set to 0.125 chips, 0.25 chips, etc., and this embodiment does not limit this.

[0122] Step 202, obtain a first delay range difference corresponding to the same amplitude range of the first measured GNSS-R normalized one-dimensional delay waveform and the first theoretical GNSS-R normalized one-dimensional delay waveform.

[0123] In an alternative embodiment, the process of obtaining a first delay range difference corresponding to the same amplitude range of the first measured GNSS-R normalized one-dimensional delay waveform and the first theoretical GNSS-R normalized one-dimensional delay waveform may include the following steps C1 to C3:

[0124] Step C1, obtain a first theoretical specular reflection point of the first theoretical GNSS-R normalized one-dimensional delay waveform, and obtain a first amplitude range corresponding to a first delay range centered on the first theoretical specular reflection point in the first theoretical GNSS-R normalized one-dimensional delay waveform.

[0125] In the embodiment of the present application, the first theoretical specular reflection point of the first theoretical GNSS-R normalized one-dimensional delay waveform refers to the point (delay point) with a delay of 0 in the first theoretical GNSS-R normalized one-dimensional delay waveform.

[0126] Exemplarily, the process of obtaining the first amplitude range corresponding to the first time delay range centered on the first theoretical specular reflection point in the first theoretical GNSS-R normalized one-dimensional time delay waveform may include: performing linear fitting on the waveform of the first time delay range centered on the first theoretical specular reflection point in the first theoretical GNSS-R normalized one-dimensional time delay waveform to obtain a first theoretical fitting waveform; obtaining the amplitude range corresponding to the first time delay range in the first theoretical fitting waveform as the first amplitude range corresponding to the first time delay range centered on the first theoretical specular reflection point in the first theoretical GNSS-R normalized one-dimensional time delay waveform.

[0127] Performing linear fitting on the waveform aims to approximately represent the trend in the waveform data through a straight line. For the process of linear fitting, those skilled in the art can handle it in any applicable manner according to actual experience, and this embodiment does not limit it. For example, it can be implemented by the least squares method, which will find a straight line that minimizes the sum of the squares of the perpendicular distances from all points on the waveform to this straight line. Another example is that in Python, the polyfit function of the NumPy library can be used for linear fitting, and so on.

[0128] For the specific value of the first time delay range, it can be set according to actual needs, and this embodiment does not limit it. For example, since the effective wave height at the specular reflection point is measured, the information at the specular reflection point mainly corresponds to the time delay range from -0.25 chips to 0.25 chips, so the first time delay range can be set as [-0.25, 0.25] chips. Of course, the first time delay range can also be set to a larger or smaller range, such as [-0.2, 0.2] chips, [-0.3, 0.3] chips, and so on.

[0129] Figure 4 It is a schematic diagram of linear fitting of a GNSS-R normalized one-dimensional time delay waveform according to an embodiment of the present application. As Figure 4 shown, where (a) represents the GNSS-R normalized one-dimensional time delay waveform before linear fitting, the curve in (b) represents the waveform between the time delay ranges [-0.25, 0.25] chips in the GNSS-R normalized one-dimensional time delay waveform in (a), and the straight line in (b) represents the fitting waveform obtained by performing linear fitting on the above curve.

[0130] In the process of obtaining the first amplitude range corresponding to the first time delay range in the first theoretical fitting waveform, obtain the amplitude (i.e., the normalized value) corresponding to the lower limit of the first time delay range in the first theoretical fitting waveform as the lower limit of the first amplitude range corresponding to the first time delay range, and obtain the amplitude (i.e., the normalized value) corresponding to the upper limit of the first time delay range in the first theoretical fitting waveform as the upper limit of the first amplitude range corresponding to the first time delay range.

[0131] Taking the first time delay range of [-0.25, 0.25] chips as an example, the first amplitude range corresponding to the first time delay range in the first theoretical fitting waveform can be obtained through the following formulas (3) and (4):

[0132]

[0133] where, represents the normalized value corresponding to a time delay of -0.25 in the first theoretical fitting waveform, represents the amplitude corresponding to a time delay of -0.25 in the first theoretical fitting waveform, represents the normalized value corresponding to a time delay of 0.25 in the first theoretical fitting waveform, represents the amplitude corresponding to a time delay of 0.25 in the first theoretical fitting waveform. Therefore, the first amplitude range corresponding to the first time delay range in the first theoretical fitting waveform is

[0134] Step C2: Obtain the first measured specular reflection point of the first measured GNSS-R normalized one-dimensional time delay waveform, and obtain the second time delay range centered on the first measured specular reflection point corresponding to the first measured GNSS-R normalized one-dimensional time delay waveform within the first amplitude range.

[0135] Due to errors such as hardware delay, the specular reflection point (i.e., the time delay 0 chip position) of the first measured GNSS-R normalized one-dimensional time delay waveform may be inaccurate. Therefore, the first theoretical GNSS-R normalized one-dimensional time delay waveform and the first measured GNSS-R normalized one-dimensional time delay waveform can be matched, the time delay offset of the first measured GNSS-R normalized one-dimensional time delay waveform can be adjusted within a certain range, and the optimal offset can be searched according to the least squares method to obtain the specular reflection point of the first measured GNSS-R normalized one-dimensional time delay waveform, which is called the first measured specular reflection point.

[0136] Exemplarily, the process of obtaining the first measured specular reflection point of the first measured GNSS-R normalized one-dimensional time delay waveform may include the following steps C21 to C23:

[0137] Step C21: Select the preset time-delay offsets one by one.

[0138] In the embodiments of the present application, a time-delay offset range can be preset according to actual needs, and multiple time-delay offsets are included in the time-delay offset range.

[0139] For example, select the part of the first measured GNSS-R normalized one-dimensional time-delay waveform with a time-delay range of [-1, 1] chips. If the time-delay sampling interval is 0.125 chips, the corresponding time-delay offset range is [-1 / 0.125, 1 / 0.125], that is, [-8, 8]. Therefore, the time-delay offsets within the time-delay offset range [-8, 8] are -8, -7, -6,..., and so on, until the time-delay offset is 8.

[0140] Step C22: Calculate the cost function shown in the following formula (5) according to each of the time-delay offsets respectively:

[0141]

[0142] where J(k) represents the cost function, represents the normalized value corresponding to the time-delay τ i+k in the first measured GNSS-R normalized one-dimensional time-delay waveform, represents the normalized value corresponding to the time-delay τ i in the first theoretical GNSS-R normalized one-dimensional time-delay waveform, k represents the time-delay offset, N1 represents the subscript of the lower limit of the time-delay in the third time-delay range, and N2 represents the subscript of the upper limit of the time-delay in the third time-delay range. Among them, the subscript represents the position of the time-delay among all the time-delay sampling points.

[0143] For the specific value of the third time-delay range, it can be set according to actual needs. For example, the third time-delay range can be set to [-4, 4] chips, [-5, 5] chips, [-6, 6] chips, etc. This embodiment does not limit this.

[0144] Step C23: Select the target time-delay offset corresponding to the minimum of the cost function, and use the time-delay corresponding to the first theoretical specular reflection point offset by the target time-delay offset as the measured specular reflection point of the first measured GNSS-R normalized one-dimensional time-delay waveform.

[0145] For example, assume that the original time-delay τ of the first theoretical GNSS-R normalized one-dimensional time-delay waveform and the first measured GNSS-R normalized one-dimensional time-delay waveform is as shown in the following formula (6):

[0146] τ = [τ1, τ2,... τ m , τ N (6)

[0147] where N is the total number of time-delay sampling points, and the time-delay sampling interval is Δτ = 0.125 chip, that is, T n+1 −τ n = Δτ.

[0148] Let the subscript of the time delay corresponding to the first theoretical specular reflection point of the first theoretical GNSS-R normalized one-dimensional time-delay waveform be m, that is, τ m = 0.

[0149] When J(k) is the smallest, m + k is the subscript of the time delay corresponding to the first measured specular reflection point of the first measured GNSS-R normalized one-dimensional time-delay waveform. At this time, the time delay τ′ of the first measured GNSS-R normalized one-dimensional time-delay waveform is adjusted to the following formula (7):

[0150] τ′ = [τ′1, τ′2,... τ′ m , τ′ N (7)

[0151] where τ′ i = τ i − kΔτ. At this time, τ′ m+k = T m+k − kΔτ = 0.

[0152] Exemplarily, the process of obtaining the second time-delay range corresponding to the first measured GNSS-R normalized one-dimensional time-delay waveform within the first amplitude range and centered on the first measured specular reflection point may include: performing linear fitting on the waveform within the first time-delay range centered on the first measured specular reflection point in the first measured GNSS-R normalized one-dimensional time-delay waveform to obtain the first measured fitting waveform; calculating the time-delay range corresponding to the first amplitude range in the first measured fitting waveform as the second time-delay range.

[0153] Performing linear fitting on the waveform within the corresponding first time-delay range τ ′ = [−0.25, 0.25] chip in the first measured GNSS-R normalized one-dimensional time-delay waveform to obtain the first measured fitting waveform

[0154] Exemplarily, the linear interpolation method can be used to calculate the time-delay range corresponding to the first amplitude range in the first measured fitting waveform through the following formulas (8) and (9):

[0155]

[0156] where represents the lower limit of the first amplitude range, Represents the lower limit of the time delay range corresponding to the first amplitude range, Represents the time delay in the first measured fitting waveform The corresponding normalized value, Represents the upper limit of the first amplitude range, Represents the upper limit of the time delay range corresponding to the first amplitude range, Represents the time delay of the first measured fitting waveform The corresponding normalized value.

[0157] Step C3, calculate the difference between the second time delay range and the first time delay range to obtain the first time delay range difference.

[0158] Exemplarily, the process of calculating the difference between the second time delay range and the first time delay range to obtain the first time delay range difference may include: calculating the first difference between the upper limit and the lower limit of the first time delay range, calculating the second difference between the upper limit and the lower limit of the second time delay range, and determining the difference between the second difference and the first difference as the first time delay range difference. The unit of the first time delay range difference is chip (specifically, the GNSS signal chip).

[0159] For example, if the first time delay range is [-0.25, 0.25] chips and the second time delay range is Then the first difference between the upper limit and the lower limit of the first time delay range is 0.25 - (-0.25) = 0.5, and the second difference between the upper limit and the lower limit of the second time delay range is Therefore, the first time delay range difference is (h 0 - 0.5).

[0160] Figure 5 Is a schematic diagram of a fitting waveform according to an embodiment of the present application. As Figure 5 Shown, where the fitting waveform with a larger slope is the first theoretical fitting waveform, and the fitting waveform with a smaller slope is the first measured fitting waveform. The first amplitude range corresponding to the first time delay range [-0.25, 0.25] chips in the first theoretical fitting waveform is The first amplitude range in the first measured fitting waveform The corresponding time delay range, that is, the second time delay range, is The first difference between the upper limit and the lower limit of the first time delay range is 0.5, and the second difference between the upper limit and the lower limit of the second time delay range is h 0 .

[0161] Step 203, perform incident angle correction on the first time delay range difference to obtain the first time delay difference observable.

[0162] Due to the sea surface wave crests and troughs caused by sea waves, the second difference is h 0It will be greater than the first difference of 0.5. There is a linear relationship between the first time delay range difference and the sea wave size, i.e., the significant wave height. Since GNSS-R is obliquely incident, the incident angle needs to be corrected.

[0163] Figure 6 This is a schematic diagram of the change in propagation distance caused by the change in sea surface height in an embodiment of the present application. As Figure 6 shown, when there are sea waves, the incident angle is θ, the sea surface height changes by h, and the overall change in propagation distance caused is approximately 2hcos(θ).

[0164] Exemplarily, the process of correcting the incident angle for the first time delay range difference to obtain the first time delay difference observation value may include: determining the ratio obtained by dividing the first time delay range difference by twice the cosine value of the first incident angle as the first time delay difference observation value. The first incident angle refers to the incident angle during the generation process of the significant wave height inversion model.

[0165] Taking the first difference between the upper limit and the lower limit of the first time delay range as 0.5, the second difference between the upper limit and the lower limit of the second time delay range as h o , and the first incident angle as θ as an example, the first time delay difference observation value Δh can be calculated by the following formula (10):

[0166]

[0167] Step 204, obtain the first significant wave height that matches the first time delay difference observation value.

[0168] Exemplarily, the process of obtaining the first significant wave height that matches the first time delay difference observation value may include: matching the first time delay difference observation value with the significant wave height measured by the buoy according to a set matching rule to obtain the first significant wave height that matches the first time delay difference observation value. Among them, the matching rule includes: the spatial distance between the specular reflection point position corresponding to the GNSS-R measurement and the buoy position corresponding to the buoy measurement is less than a preset distance, and the time interval between the observation time corresponding to the GNSS-R measurement and the observation time corresponding to the buoy measurement is less than a preset duration.

[0169] For the specific values of the above preset distance and preset duration, they can be set according to actual needs. For example, the preset distance can be set to 25 km, the preset duration can be set to 1 hour, etc. This embodiment does not limit this.

[0170] During the GNSS-R measurement process, a first time delay difference observation quantity can be calculated at each preset time interval, and then a corresponding first significant wave height can be obtained by matching. Multiple GNSS-R data (such as GNSS-R data over 1 month, etc.) are used for matching to obtain a matching data set of the first time delay difference observation quantity measured by GNSS-R and the first significant wave height measured by the buoy.

[0171] Step 205: Use the first time delay difference observation quantity and the first significant wave height to train an effective wave height inversion model.

[0172] Exemplarily, the process of using the first time delay difference observation quantity and the first significant wave height to train an effective wave height inversion model may include: performing linear fitting training using the first time delay difference observation quantity and the first significant wave height (specifically the above-mentioned matching data set) to obtain an effective wave height inversion model as shown in the following formula (11):

[0173] H s =aΔh + b (11)

[0174] Where, H s represents the significant wave height, Δh represents the time delay difference observation quantity, a represents the first coefficient, and b represents the second coefficient.

[0175] The goal of linear fitting is to find a best straight line that can best describe the relationship between data points (the first time delay difference observation quantity and the first significant wave height). For example, it can be achieved by minimizing the sum of the squares of the perpendicular distances from the data points to the fitting straight line (i.e., the least squares method), etc. After the linear fitting training is completed, the first coefficient a and the second coefficient b of the effective wave height inversion model as shown in the above formula (11) can be obtained.

[0176] In the embodiments of the present application, an effective wave height inversion based on GNSS-R technology is proposed. The original waveform is normalized to obtain a GNSS-R normalized one-dimensional time delay waveform. The time delay information rather than the power amplitude information of the GNSS-R normalized one-dimensional time delay waveform is used for comparative analysis to obtain a first time delay difference observation quantity. Then, an effective wave height inversion model is trained based on the first time delay difference observation quantity and the matched first significant wave height. Using time delay information analysis compared with using power amplitude information analysis, there is no need to calibrate the signal power, which is more direct and has higher accuracy. There are no problems such as a decrease in sensitivity under high sea conditions and a reduction in accuracy when the swell dominates. And when calculating the first time delay difference observation quantity, an incident angle correction is added, further improving the accuracy of the first time delay difference observation quantity, thereby improving the accuracy of the effective wave height inversion model, and further improving the accuracy of the significant wave height inverted by the effective wave height inversion model.

[0177] Refer to Figure 7, which shows the step flowchart of an effective wave height inversion method according to an embodiment of the present application.

[0178] As Figure 7 shown, the effective wave height inversion method may include the following steps:

[0179] Step 701, obtain the second measured GNSS-R normalized one-dimensional delay waveform and obtain the second theoretical GNSS-R normalized one-dimensional delay waveform.

[0180] In the embodiment of the present application, the second measured GNSS-R normalized one-dimensional delay waveform refers to the measured GNSS-R normalized one-dimensional delay waveform used in the effective wave height inversion process, and the second theoretical GNSS-R normalized one-dimensional delay waveform refers to the theoretical GNSS-R normalized one-dimensional delay waveform used in the effective wave height inversion process. Therefore, the second measured GNSS-R normalized one-dimensional delay waveform and the second theoretical GNSS-R normalized one-dimensional delay waveform can be referred to as data to be processed.

[0181] Exemplarily, obtaining the second measured GNSS-R normalized one-dimensional delay waveform includes: obtaining the second measured DDM waveform; performing denoising processing on the second measured DDM waveform to obtain a second target measured DDM waveform; extracting the value corresponding to a Doppler frequency of 0 in the second target measured DDM waveform as a third target value, and dividing the third target value by the peak value in the third target value to obtain the second measured GNSS-R normalized one-dimensional delay waveform.

[0182] Exemplarily, the performing denoising processing on the second measured DDM waveform to obtain a second target measured DDM waveform includes: calculating a third average value of the values corresponding to the delays less than or equal to a preset delay threshold in the second measured DDM waveform, and using the third average value as the noise floor of the second measured DDM waveform; subtracting the noise floor of the second measured DDM waveform from the second measured DDM waveform to obtain the second target measured DDM waveform.

[0183] Optionally, obtaining the second theoretical GNSS-R normalized one-dimensional delay waveform includes: obtaining the second theoretical DDM waveform; performing denoising processing on the second theoretical DDM waveform to obtain a second target theoretical DDM waveform; extracting the value corresponding to a Doppler frequency of 0 in the second target theoretical DDM waveform as a fourth target value, and dividing the fourth target value by the peak value in the fourth target value to obtain the second theoretical GNSS-R normalized one-dimensional delay waveform.

[0184] Optionally, the denoising process for the second theoretical DDM waveform to obtain the second target theoretical DDM waveform includes: calculating a fourth average value of the values corresponding to the time delays less than or equal to a preset time delay threshold in the second theoretical DDM waveform, and using the fourth average value as the noise floor of the second theoretical DDM waveform; subtracting the noise floor of the second theoretical DDM waveform from the second theoretical DDM waveform to obtain the second target theoretical DDM waveform.

[0185] Step 702, obtain a second time delay range difference corresponding to the same amplitude range of the second measured GNSS-R normalized one-dimensional time delay waveform and the second theoretical GNSS-R normalized one-dimensional time delay waveform.

[0186] Exemplarily, obtaining the second time delay range difference corresponding to the same amplitude range of the second measured GNSS-R normalized one-dimensional time delay waveform and the second theoretical GNSS-R normalized one-dimensional time delay waveform includes: obtaining a second theoretical specular reflection point of the second theoretical GNSS-R normalized one-dimensional time delay waveform, and obtaining a second amplitude range corresponding to a first time delay range centered on the second theoretical specular reflection point in the second theoretical GNSS-R normalized one-dimensional time delay waveform; obtaining a second measured specular reflection point of the second measured GNSS-R normalized one-dimensional time delay waveform, and obtaining a fourth time delay range centered on the second measured specular reflection point corresponding to the second measured GNSS-R normalized one-dimensional time delay waveform under the second amplitude range; calculating a difference between the fourth time delay range and the first time delay range to obtain the second time delay range difference.

[0187] Exemplarily, obtaining the second amplitude range corresponding to the first time delay range centered on the second theoretical specular reflection point in the second theoretical GNSS-R normalized one-dimensional time delay waveform includes: performing linear fitting on the waveform of the first time delay range centered on the second theoretical specular reflection point in the second theoretical GNSS-R normalized one-dimensional time delay waveform to obtain a second theoretical fitting waveform; obtaining the amplitude range corresponding to the first time delay range in the second theoretical fitting waveform as the second amplitude range.

[0188] For the process of obtaining the second measured specular reflection point of the second measured GNSS-R normalized one-dimensional time delay waveform, reference may be made to the relevant description of the process of obtaining the first measured specular reflection point of the first measured GNSS-R normalized one-dimensional time delay waveform above.

[0189] Exemplarily, the obtaining of the fourth time delay range corresponding to the second measured GNSS-R normalized one-dimensional time delay waveform within the second amplitude range and centered on the second measured specular reflection point includes: linearly fitting the waveform within the first time delay range centered on the second measured specular reflection point in the second measured GNSS-R normalized one-dimensional time delay waveform to obtain a second measured fitted waveform; calculating the time delay range corresponding to the second amplitude range in the second measured fitted waveform as the fourth time delay range.

[0190] Step 703: Perform incident angle correction on the second time delay range difference to obtain a second time delay difference observation.

[0191] Optionally, the performing of incident angle correction on the second time delay range difference to obtain a second time delay difference observation includes: determining the ratio obtained by dividing the second time delay range difference by twice the cosine value of the second incident angle as the second time delay difference observation. The second incident angle refers to the incident angle during the effective wave height inversion process.

[0192] Step 704: Use a pre-trained effective wave height inversion model to calculate the effective wave height corresponding to the second time delay difference observation.

[0193] After obtaining the second time delay difference observation, the effective wave height inversion model shown in formula (11) above can be used to substitute the second time delay difference observation as Δh into the effective wave height inversion model, thereby calculating the effective wave height H corresponding to the second time delay difference observation. s 。

[0194] In the embodiments of the present application, the accuracy of the effective wave height inverted by the effective wave height inversion model can be improved.

[0195] Refer to Figure 8 , which shows a structural block diagram of an effective wave height inversion model generation device according to an embodiment of the present application.

[0196] As Figure 8 shown, the effective wave height inversion model generation device may include the following modules:

[0197] The first acquisition module 801 is configured to acquire a first measured GNSS-R normalized one-dimensional time delay waveform and acquire a first theoretical GNSS-R normalized one-dimensional time delay waveform;

[0198] The second acquisition module 802 is configured to acquire a first time delay range difference corresponding to the same amplitude range of the first measured GNSS-R normalized one-dimensional time delay waveform and the first theoretical GNSS-R normalized one-dimensional time delay waveform;

[0199] The first correction module 803 is configured to correct the incident angle of the first time-delay range difference to obtain a first time-delay difference observation value;

[0200] The matching module 804 is configured to obtain a first effective wave height that matches the first time-delay difference observation value;

[0201] The training module 805 is configured to train an effective wave height inversion model by using the first time-delay difference observation value and the first effective wave height.

[0202] Optionally, the first acquisition module 801 includes: a first waveform acquisition sub-module, configured to acquire a first measured DDM waveform; a first denoising sub-module, configured to perform denoising processing on the first measured DDM waveform to obtain a first target measured DDM waveform; a first normalization sub-module, configured to extract a value corresponding to a Doppler frequency of 0 in the first target measured DDM waveform as a first target value, and divide the first target value by a peak value in the first target value to obtain the first measured GNSS-R normalized one-dimensional time-delay waveform.

[0203] Optionally, the first denoising sub-module is specifically configured to calculate a first average value of values corresponding to time delays less than or equal to a preset time-delay threshold in the first measured DDM waveform, and use the first average value as the noise floor of the first measured DDM waveform; subtract the noise floor of the first measured DDM waveform from the first measured DDM waveform to obtain the first target measured DDM waveform.

[0204] Optionally, the first acquisition module 801 includes: a second waveform acquisition sub-module, configured to acquire a first theoretical DDM waveform; a second denoising sub-module, configured to perform denoising processing on the first theoretical DDM waveform to obtain a first target theoretical DDM waveform; a second normalization sub-module, configured to extract a value corresponding to a Doppler frequency of 0 in the first target theoretical DDM waveform as a second target value, and divide the second target value by a peak value in the second target value to obtain the first theoretical GNSS-R normalized one-dimensional time-delay waveform.

[0205] Optionally, the second denoising sub-module is specifically configured to calculate a second average value of values corresponding to time delays less than or equal to a preset time-delay threshold in the first theoretical DDM waveform, and use the second average value as the noise floor of the first theoretical DDM waveform; subtract the noise floor of the first theoretical DDM waveform from the first theoretical DDM waveform to obtain the first target theoretical DDM waveform.

[0206] Optionally, the second acquisition module 802 includes: a first range acquisition sub-module, configured to acquire a first theoretical specular reflection point of the first theoretical GNSS-R normalized one-dimensional time-delay waveform, and acquire a first amplitude range corresponding to a first time-delay range centered on the first theoretical specular reflection point in the first theoretical GNSS-R normalized one-dimensional time-delay waveform; a second range acquisition sub-module, configured to acquire a first measured specular reflection point of the first measured GNSS-R normalized one-dimensional time-delay waveform, and acquire a second time-delay range corresponding to the first measured GNSS-R normalized one-dimensional time-delay waveform under the first amplitude range and centered on the first measured specular reflection point; a calculation sub-module, configured to calculate a difference between the second time-delay range and the first time-delay range to obtain the first time-delay range difference.

[0207] Optionally, the first range acquisition sub-module includes: a first fitting unit, configured to perform linear fitting on a waveform of a first time-delay range centered on the first theoretical specular reflection point in the first theoretical GNSS-R normalized one-dimensional time-delay waveform to obtain a first theoretical fitting waveform; and acquire an amplitude range corresponding to the first time-delay range in the first theoretical fitting waveform as the first amplitude range.

[0208] Optionally, the second range acquisition sub-module includes a reflection point acquisition unit, and the reflection point acquisition unit is configured to:

[0209] select a preset time-delay offset one by one;

[0210] calculate the following cost function according to each of the time-delay offsets respectively:

[0211]

[0212] where J(k) represents the cost function, represents the normalized value corresponding to the time delay τ in the first measured GNSS-R normalized one-dimensional time-delay waveform i+k corresponding thereto, represents the normalized value corresponding to the time delay τ in the first theoretical GNSS-R normalized one-dimensional time-delay waveform i corresponding thereto, k represents the time-delay offset, N1 represents the subscript of the lower limit of the time delay of the third time-delay range, and N2 represents the subscript of the upper limit of the time delay of the third time-delay range;

[0213] select the target time-delay offset corresponding to the minimum value of the cost function, and use the time delay corresponding to the first theoretical specular reflection point offset by the target time-delay offset as the measured specular reflection point of the first measured GNSS-R normalized one-dimensional time-delay waveform.

[0214] Optionally, the first range acquisition sub-module includes: a second fitting unit, configured to perform linear fitting on the waveform within a first time delay range centered on the first measured specular reflection point in the first measured GNSS-R normalized one-dimensional time delay waveform, to obtain a first measured fitted waveform; and calculate the time delay range corresponding to the first amplitude range in the first measured fitted waveform as the second time delay range.

[0215] Optionally, the first correction module 803 is specifically configured to determine the ratio obtained by dividing the first time delay range difference by twice the cosine value of the first incident angle as the first time delay difference observation.

[0216] Optionally, the matching module 804 is specifically configured to match the first time delay difference observation with the significant wave height measured by the buoy according to a set matching rule, to obtain a first significant wave height that matches the first time delay difference observation; the matching rule includes: the spatial distance between the position of the specular reflection point corresponding to the GNSS-R measurement and the position of the buoy corresponding to the buoy measurement is less than a preset distance, and the time interval between the observation time corresponding to the GNSS-R measurement and the observation time corresponding to the buoy measurement is less than a preset duration.

[0217] Optionally, the training module 805 is specifically configured to perform linear fitting training using the first time delay difference observation and the first significant wave height, to obtain the following significant wave height inversion model:

[0218] H s = aΔh + b

[0219] where H s represents the significant wave height, Δh represents the time delay difference observation, a represents the first coefficient, and b represents the second coefficient.

[0220] Refer to Figure 9 , which shows the structural block diagram of a significant wave height inversion device according to an embodiment of the present application.

[0221] As Figure 9 shown, the significant wave height inversion device may include the following modules:

[0222] A third acquisition module 901, configured to acquire a second measured GNSS-R normalized one-dimensional time delay waveform and acquire a second theoretical GNSS-R normalized one-dimensional time delay waveform;

[0223] A fourth acquisition module 902, configured to acquire a second time delay range difference corresponding to the same amplitude range of the second measured GNSS-R normalized one-dimensional time delay waveform and the second theoretical GNSS-R normalized one-dimensional time delay waveform;

[0224] A second correction module 903, configured to perform incident angle correction on the second time delay range difference to obtain a second time delay difference observation value;

[0225] A calculation module 904, configured to calculate an effective wave height corresponding to the second time delay difference observation value by using a pre-trained effective wave height inversion model; the effective wave height inversion model is obtained by the effective wave height inversion model generation method described in any one of the above.

[0226] For the apparatus embodiment, since it is basically similar to the method embodiment, the description is relatively simple. For the related parts, please refer to the partial description of the method embodiment.

[0227] In an embodiment of the present application, an electronic device is further provided. The electronic device may include a processor and a computer-readable storage medium, and a computer program is stored on the computer-readable storage medium; when the computer program is executed by the processor, the processor executes the effective wave height inversion model generation method in any one of the above embodiments, or executes the effective wave height inversion method in any one of the above embodiments.

[0228] Refer to Figure 10 , which shows a structural block diagram of an electronic device according to an embodiment of the present application. As Figure 10 shown, the electronic device 100 includes a processor 1001 and a computer-readable storage medium 1002, and a computer program 10021 is stored on the computer-readable storage medium 1002.

[0229] The processor 1001 is configured to execute the computer program 10021 stored on the computer-readable storage medium 1002. When the processor 1001 executes the computer program 10021, it executes the effective wave height inversion model generation method in any one of the above embodiments, or executes the effective wave height inversion method in any one of the above embodiments, and can achieve the same technical effect. To avoid repetition, it will not be described in detail here.

[0230] The above-mentioned processor 1001 may include, but is not limited to: a central processing unit (CPU for short), a network processor (NP for short), a digital signal processor (DSP for short), an application specific integrated circuit (ASIC for short), a field-programmable gate array (FPGA for short), or other programmable logic devices, discrete gate or transistor logic devices, discrete hardware components, and so on.

[0231] The aforementioned computer-readable storage medium 1002 may include, but is not limited to: Read Only Memory (ROM), Random Access Memory (RAM), Compact Disc Read Only Memory (CD-ROM), Electronic Erasable Programmable Read Only Memory (EEPROM), hard disk, floppy disk, flash memory, and so on.

[0232] In an embodiment of the present application, a computer-readable storage medium is further provided. A computer program is stored on the computer-readable storage medium, and the computer program can be executed by a processor of an electronic device. When the computer program is executed by the processor, the processor is caused to execute the effective wave height inversion model generation method of any of the above embodiments, or execute the effective wave height inversion method of any of the above embodiments.

[0233] Refer to Figure 11 , which shows a structural block diagram of a computer-readable storage medium according to an embodiment of the present application. As Figure 11 shown, a computer program 1101 is stored on the computer-readable storage medium 110. When the computer program 1101 is executed by the processor, the processor is caused to execute the effective wave height inversion model generation method of any of the above embodiments, or execute the effective wave height inversion method of any of the above embodiments, and the same technical effects can be achieved. To avoid repetition, it will not be elaborated here.

[0234] The various embodiments in this specification are interrelated. Each embodiment is described in a progressive manner. The key point of each embodiment is to illustrate the differences from other embodiments. For the same or similar parts among the various embodiments, reference can be made to each other.

[0235] It should be noted that in this article, relational terms such as first and second are only used to distinguish one entity or operation from another entity or operation, and do not necessarily require or imply any actual relationship or order between these entities or operations. Moreover, the term "comprising", "including" or any other variant thereof is intended to cover non-exclusive inclusion, so that a process, method, article or terminal device including a series of elements not only includes those elements, but also includes other elements not explicitly listed, or further includes elements inherent to such process, method, article or terminal device. Without further limitation, an element defined by the statement "including one..." does not exclude the existence of another identical element in the process, method, article or terminal device including the said element.

[0236] Through the description of the above embodiments, those skilled in the art can clearly understand that the above-described method of the embodiments can be implemented by means of software plus a necessary general hardware platform. Of course, it can also be implemented by hardware, but in many cases the former is a better implementation. Based on such an understanding, the technical solution of the present application, in essence, or the part that contributes to the prior art, or a part of the technical solution, can be embodied in the form of a software product. The computer software product is stored in a storage medium and includes several instructions for causing a computer device (which can be a personal computer, a server, or a network device, etc.) to execute all or part of the steps of the methods described in the various embodiments of the present application.

[0237] The embodiments of the present application have been described above in conjunction with the accompanying drawings. However, the present application is not limited to the above specific implementation manners. The above specific implementation manners are merely illustrative and not restrictive. Under the inspiration of the present application, those of ordinary skill in the art can also make many forms without departing from the purpose of the present application and the scope protected by the claims, and all of them belong to the protection scope of the present application.

[0238] Those of ordinary skill in the art can realize that the units and algorithm steps of each example described in combination with the embodiments disclosed in the embodiments of the present application can be implemented by electronic hardware, or a combination of computer software and electronic hardware. Whether these functions are executed in a hardware or software manner depends on the specific application and design constraints of the technical solution. Professional technicians can use different methods to implement the described functions for each specific application, but such implementation should not be considered to exceed the scope of the present application.

[0239] Those skilled in the art can clearly understand that for the convenience and conciseness of description, the specific working processes of the above-described systems, devices, and units can refer to the corresponding processes in the foregoing method embodiments and will not be repeated here.

[0240] In the embodiments provided by the present application, it should be understood that the disclosed devices and methods can be implemented in other ways. For example, the device embodiments described above are merely illustrative. For example, the division of the units is only a logical function division. In actual implementation, there may be other division methods. For example, multiple units or components can be combined or integrated into another system, or some features can be ignored or not executed. Another point is that the displayed or discussed couplings or direct couplings or communication connections to each other can be through some interfaces. The indirect couplings or communication connections of the devices or units can be in an electrical, mechanical, or other form.

[0241] The unit described as a separation component may or may not be physically separated. The component shown as a unit may or may not be a physical unit, that is, it may be located in one place or may be distributed across multiple network units. Some or all of the units can be selected according to actual needs to achieve the purpose of the solution of this embodiment.

[0242] In addition, in each embodiment of the present application, each functional unit may be integrated in a processing unit, may exist separately as individual physical units, or two or more units may be integrated in one unit.

[0243] As described above, the above are only specific embodiments of the present application, but the protection scope of the present application is not limited thereto. Any person skilled in the art within the technical scope disclosed in the present application can easily think of changes or substitutions, which should all be covered within the protection scope of the present application. In summary, the content of this specification should not be construed as a limitation to the present application.

Claims

1. A method for generating an effective wave height inversion model, characterized in that, The method includes: Obtaining a first measured GNSS-R normalized one-dimensional delay waveform and obtaining a first theoretical GNSS-R normalized one-dimensional delay waveform; Obtaining a first delay range difference corresponding to the same amplitude range of the first measured GNSS-R normalized one-dimensional delay waveform and the first theoretical GNSS-R normalized one-dimensional delay waveform; Performing incidence angle correction on the first delay range difference to obtain a first delay difference observation; Obtaining a first effective wave height that matches the first delay difference observation; Training an effective wave height inversion model using the first delay difference observation and the first effective wave height; The obtaining the first delay range difference corresponding to the same amplitude range of the first measured GNSS-R normalized one-dimensional delay waveform and the first theoretical GNSS-R normalized one-dimensional delay waveform includes: Obtaining a first theoretical specular reflection point of the first theoretical GNSS-R normalized one-dimensional delay waveform and obtaining a first amplitude range corresponding to a first delay range centered on the first theoretical specular reflection point in the first theoretical GNSS-R normalized one-dimensional delay waveform; Obtaining a first measured specular reflection point of the first measured GNSS-R normalized one-dimensional delay waveform and obtaining a second delay range centered on the first measured specular reflection point corresponding to the first measured GNSS-R normalized one-dimensional delay waveform under the first amplitude range; Calculating the difference between the second delay range and the first delay range to obtain the first delay range difference.

2. The method according to claim 1, characterized in that, Obtaining a first measured GNSS-R normalized one-dimensional delay waveform includes: Obtaining a first measured DDM waveform; Performing denoising processing on the first measured DDM waveform to obtain a first target measured DDM waveform; Extracting a value corresponding to a Doppler frequency of 0 in the first target measured DDM waveform as a first target value, and dividing the first target value by a peak value in the first target value to obtain the first measured GNSS-R normalized one-dimensional delay waveform.

3. The method according to claim 2, wherein The performing denoising processing on the first measured DDM waveform to obtain a first target measured DDM waveform includes: Calculating a first average value of values corresponding to delays less than or equal to a preset delay threshold in the first measured DDM waveform, and using the first average value as the noise floor of the first measured DDM waveform; Subtracting the noise floor of the first measured DDM waveform from the first measured DDM waveform to obtain the first target measured DDM waveform.

4. The method according to claim 1, characterized in that, Obtaining a first theoretical GNSS-R normalized one-dimensional delay waveform includes: Obtaining a first theoretical DDM waveform; Performing denoising processing on the first theoretical DDM waveform to obtain a first target theoretical DDM waveform; Extracting a value corresponding to a Doppler frequency of 0 in the first target theoretical DDM waveform as a second target value, and dividing the second target value by a peak value in the second target value to obtain the first theoretical GNSS-R normalized one-dimensional delay waveform.

5. The method according to claim 4, characterized in that The performing denoising processing on the first theoretical DDM waveform to obtain a first target theoretical DDM waveform includes: Calculate the second average value of the values corresponding to the time delays less than or equal to the preset time delay threshold in the first theoretical DDM waveform, and use the second average value as the noise floor of the first theoretical DDM waveform. Subtract the noise floor of the first theoretical DDM waveform from the first theoretical DDM waveform to obtain the first target theoretical DDM waveform.

6. The method according to claim 1, characterized in that The obtaining of the first amplitude range corresponding to the first time delay range centered on the first theoretical specular reflection point in the first theoretical GNSS-R normalized one-dimensional time delay waveform includes: Perform linear fitting on the waveform of the first time delay range centered on the first theoretical specular reflection point in the first theoretical GNSS-R normalized one-dimensional time delay waveform to obtain a first theoretical fitting waveform. Obtain the amplitude range corresponding to the first time delay range in the first theoretical fitting waveform as the first amplitude range.

7. The method according to claim 1, characterized in that, The obtaining of the first measured specular reflection point of the first measured GNSS-R normalized one-dimensional time delay waveform includes: Select the preset time delay offsets one by one. Calculate the following cost function according to each of the time delay offsets respectively: Wherein, J(k) represents the cost function, represents the time delay τ in the first measured GNSS-R normalized one-dimensional time delay waveform i+k The corresponding normalized value is represents the time delay τ in the first theoretical GNSS-R normalized one-dimensional time delay waveform i The corresponding normalized values, k represents the delay offset, N1 represents the subscript of the delay lower limit of the third delay range, and N2 represents the subscript of the delay upper limit of the third delay range; Select the target time delay offset corresponding to the minimum of the cost function, and use the time delay corresponding to the first theoretical specular reflection point offset by the target time delay offset as the measured specular reflection point of the first measured GNSS-R normalized one-dimensional time delay waveform.

8. The method according to claim 1, wherein The obtaining of the second time delay range corresponding to the first measured GNSS-R normalized one-dimensional time delay waveform in the first amplitude range and centered on the first measured specular reflection point includes: Perform linear fitting on the waveform of the first time delay range centered on the first measured specular reflection point in the first measured GNSS-R normalized one-dimensional time delay waveform to obtain a first measured fitting waveform. Calculate the time delay range corresponding to the first amplitude range in the first measured fitting waveform as the second time delay range.

9. The method according to claim 1, wherein The correcting the first time delay range difference for the incident angle to obtain the first time delay difference observation value includes: Determine the ratio obtained by dividing the first time delay range difference by twice the cosine value of the first incident angle as the first time delay difference observation value.

10. The method according to claim 1, characterized in that, The obtaining of the first effective wave height matching the first time delay difference observation value includes: Match the first time delay difference observation value with the effective wave height measured by the buoy according to the set matching rule to obtain the first effective wave height matching the first time delay difference observation value. The matching rule includes: the spatial distance between the position of the specular reflection point corresponding to the GNSS-R measurement and the position of the buoy corresponding to the buoy measurement is less than the preset distance, and the time interval between the observation time corresponding to the GNSS-R measurement and the observation time corresponding to the buoy measurement is less than the preset duration.

11. The method according to claim 1, characterized in that, The training of the effective wave height inversion model using the first time delay difference observation value and the first effective wave height includes: Perform linear fitting training using the first time delay difference observation value and the first effective wave height to obtain the following effective wave height inversion model: H s = aΔh + b Among them, H s represents the significant wave height, Δh represents the time-delay difference observation, a represents the first coefficient, and b represents the second coefficient.

12. An effective wave height inversion method, characterized in that, The method includes: Obtain the second measured GNSS-R normalized one-dimensional delay waveform and obtain the second theoretical GNSS-R normalized one-dimensional delay waveform; Obtain the second delay range difference corresponding to the same amplitude range of the second measured GNSS-R normalized one-dimensional delay waveform and the second theoretical GNSS-R normalized one-dimensional delay waveform; Perform incident angle correction on the second delay range difference to obtain the second delay difference observation; Use the pre-trained significant wave height inversion model to calculate the significant wave height corresponding to the second delay difference observation; the significant wave height inversion model is obtained by the significant wave height inversion model generation method according to any one of claims 1 to 11.

13. An effective wave height inversion model generation device, characterized in that The device includes: The first acquisition module is used to obtain the first measured GNSS-R normalized one-dimensional delay waveform and obtain the first theoretical GNSS-R normalized one-dimensional delay waveform; The second acquisition module is used to obtain the first delay range difference corresponding to the same amplitude range of the first measured GNSS-R normalized one-dimensional delay waveform and the first theoretical GNSS-R normalized one-dimensional delay waveform; The first correction module is used to perform incident angle correction on the first delay range difference to obtain the first delay difference observation; The matching module is used to obtain the first significant wave height that matches the first delay difference observation; The training module is used to train the significant wave height inversion model by using the first delay difference observation and the first significant wave height; Among them, the second acquisition module includes: a first range acquisition sub-module, which is used to obtain the first theoretical specular reflection point of the first theoretical GNSS-R normalized one-dimensional delay waveform and obtain the first amplitude range corresponding to the first delay range centered on the first theoretical specular reflection point in the first theoretical GNSS-R normalized one-dimensional delay waveform; a second range acquisition sub-module, which is used to obtain the first measured specular reflection point of the first measured GNSS-R normalized one-dimensional delay waveform and obtain the second delay range centered on the first measured specular reflection point corresponding to the first measured GNSS-R normalized one-dimensional delay waveform under the first amplitude range; a calculation sub-module, which is used to calculate the difference between the second delay range and the first delay range to obtain the first delay range difference.

14. An effective wave height inversion device, characterized in that, The device includes: The third acquisition module is used to obtain the second measured GNSS-R normalized one-dimensional delay waveform and obtain the second theoretical GNSS-R normalized one-dimensional delay waveform; The fourth acquisition module is used to obtain the second delay range difference corresponding to the same amplitude range of the second measured GNSS-R normalized one-dimensional delay waveform and the second theoretical GNSS-R normalized one-dimensional delay waveform; The second correction module is used to perform incident angle correction on the second delay range difference to obtain the second delay difference observation; The calculation module is used to use the pre-trained significant wave height inversion model to calculate the significant wave height corresponding to the second delay difference observation; the significant wave height inversion model is obtained by the significant wave height inversion model generation method according to any one of claims 1 to 11.

15. An electronic device, characterized in that, The electronic device includes a processor and a computer-readable storage medium, and a computer program is stored on the computer-readable storage medium; When the computer program is executed by the processor, the processor is caused to execute the significant wave height inversion model generation method according to any one of claims 1 to 11, or execute the significant wave height inversion method according to claim 12.

16. A computer-readable storage medium, characterized in that, A computer program is stored on the computer-readable storage medium. When the computer program is executed by the processor, the processor is caused to execute the significant wave height inversion model generation method according to any one of claims 1 to 11, or execute the significant wave height inversion method according to claim 12.

Images