Transition zone disturbance-resistant optimal incidence angle retrieval method based on small slope approximation

Abstract

The application provides a transition zone anti-disturbance optimal incidence angle retrieval method based on a small slope approximation method, belongs to the technical field of electromagnetic scattering characteristics of sea surface, obtains backscattering data of a satellite observed ocean area and corresponding radar parameters, and obtains wind field reference information of the ocean area based on buoy measurement data, and constructs a matching data set for method verification; a small slope approximation model is used as a forward model, a cost function is constructed and iterative optimization is performed, and a reference wind speed and a reference wind direction of the ocean area are obtained by inversion; a group of wind speed disturbance samples are generated in the neighborhood of the reference wind speed; the reference wind speed, the wind speed disturbance sample, the reference wind direction and the radar parameter are input into the small slope approximation model, an incidence angle corresponding to the minimum overall deviation amount is determined, and the incidence angle is used as the optimal incidence angle with the anti-disturbance ability under the current sea condition, and the accuracy and effectiveness of the optimal incidence angle retrieval method are verified.

Description

TECHNICAL FIELD

[0001] The application belongs to the field of sea surface electromagnetic scattering characteristics, and particularly relates to a transition zone anti-disturbance optimal incidence angle retrieval method based on a small slope approximation method. BACKGROUND

[0002] In the incidence angle range (about 0°~30°), the sea surface electromagnetic scattering mechanism is complex, and multiple scattering mechanisms coexist. In the incidence angle range close to 0°, the main scattering mechanism of the sea surface is quasi-specular scattering; with the increase of the incidence angle, the contribution of quasi-specular scattering gradually decreases, the contribution of Bragg scattering gradually increases, and the phenomenon of quasi-specular scattering and Bragg scattering mixed transition occurs. The incidence angle range in which the main scattering mechanism dynamically changes is called the "transition zone".

[0003] In the transition zone, the response characteristics of quasi-specular scattering and Bragg scattering to the sea surface roughness are significantly different. With the increase of the sea surface roughness, the contribution of quasi-specular scattering gradually decreases, the contribution of Bragg scattering gradually increases, and finally reaches a saturation state or is affected by wave breaking. The response difference to the change of the sea surface roughness significantly reduces the sensitivity and correlation of the sea surface electromagnetic scattering to the change of environmental factors in the transition zone. Under the condition of fixed sea surface roughness, when the incidence angle makes the contribution of quasi-specular scattering and the contribution of Bragg scattering reach a balance, the stability of the sea surface electromagnetic scattering reaches a peak, and is almost not affected by the mutation of environmental factors. The incidence angle is recorded as the optimal incidence angle.

[0004] The stability characteristics of the sea surface electromagnetic scattering in the transition zone have significant advantages in real applications, mainly reflected in strong anti-interference ability and stable observation ability. Maintaining the low sensitivity of the sea surface electromagnetic scattering to the change of environmental factors can significantly reduce the influence of environmental disturbances (such as wind speed mutation and wave change) on the sea surface electromagnetic scattering, and effectively improve the quality of observation data. At the same time, the low sensitivity of the sea surface electromagnetic scattering to the change of environmental factors helps to better distinguish the scattering signals of sea clutter and sea targets, and can effectively improve the detection ability of sea targets and reduce the false alarm rate and the missing detection rate in target detection. However, the dynamic balance of the scattering mechanism contribution in the transition zone is affected by the radar incidence angle and the sea surface roughness, and it is difficult to quickly and accurately determine the optimal incidence angle under different sea surface roughness conditions through conventional observation methods. SUMMARY

[0005] In view of the deficiencies in the prior art, the application provides a transition zone anti-disturbance optimal incidence angle retrieval method based on a small slope approximation method, which comprises the following steps:

[0006] S1. Acquire satellite observation data and buoy measurement data. The satellite observation data includes backscatter data of the ocean area and corresponding radar parameters; the buoy measurement data includes wind field information of the ocean area. Perform data preprocessing on the satellite observation data and buoy measurement data to construct a matching dataset for method verification. S2. Based on the backscatter data and corresponding radar parameters, use a small-slope approximation model as the forward model, construct a cost function and perform iterative optimization to invert and obtain the reference wind speed and reference wind direction of the ocean area. S3. Generate a range of wind speeds within the neighborhood of the reference wind speed. A set of wind speed disturbance samples is generated; the reference wind speed, the wind speed disturbance samples, the reference wind direction, and the radar parameters are input into the small slope approximation model, and the incident angle corresponding to the minimum overall deviation is retrieved as the incident angle with the best anti-disturbance capability under the current sea state; S4, based on the matching dataset constructed in step S1, the Pearson correlation coefficient between the backscattered data of the sea surface and the wind speed under different incident angles is calculated, and the accuracy and effectiveness of the optimal incident angle retrieval method are verified by comparing the zero point of the Pearson correlation coefficient with the optimal incident angle obtained in step S3.

[0007] Furthermore, in step S2, wind vector inversion is performed using the maximum likelihood estimation method, and the cost function J is:

[0008] ;

[0009] In the formula, Indicates the number of observations. The variance of the noise for the i-th observation is represented by the following expression: Indicates the first One observation value, and These represent wind speed and relative wind direction, respectively. This represents the simulation results of backscattering.

[0010] Further, step S3 includes:

[0011] (1) Based on the wind speed inversion results obtained by wind vector inversion, a set of equally spaced wind speed disturbance samples are generated in the neighborhood of the reference wind speed.

[0012] (2) Input the wind speed disturbance sample, the reference wind speed, the inverted wind direction and the radar system parameters into the small slope approximation model, and calculate the reference wind speed and the sea surface backscattering simulation data corresponding to each wind speed disturbance sample respectively.

[0013] (3) For each incident angle, calculate the weighted sum of squares of the deviations between the simulation results corresponding to each wind speed disturbance sample and the simulation results of the reference wind speed;

[0014] (4) According to the change trend of the weighted sum of squares of deviations with the incident angle, an incident angle that makes the weighted sum of squares of deviations minimum is selected as the optimal incident angle with respect to the environmental disturbance resistance.

[0015] Further, a weight coefficient is calculated according to the difference between each wind speed disturbance sample and the reference wind speed, and the weighted sum of squares of deviations of the sea surface backscattering simulation data under the wind speed sample and the reference wind speed is:

[0016] ;

[0017] In the formula, indicates the number of samples under the same incident angle condition, indicates the sea surface backscattering simulation result calculated by the jth sample wind speed, indicates the sea surface backscattering simulation result calculated by the reference wind speed, indicates the jth weight coefficient.

[0018] Further, the step S4 comprises:

[0019] (1) Extracting the sea surface backscattering data, radar parameters and wind field information from the obtained matching data set;

[0020] (2) Using the wind vector inversion method to perform wind vector inversion to obtain the reference wind speed and wind direction;

[0021] (3) Using the optimal incident angle retrieval method to determine the optimal incident angle;

[0022] (4) Under different incident angle conditions, calculating the Pearson correlation coefficient between the sea surface backscattering data and the wind speed in the measured data;

[0023] (5) Taking the Pearson correlation coefficient as an evaluation index, quantifying the sensitivity of the sea surface backscattering data to the wind speed, and comparing the difference between the zero point of the Pearson correlation coefficient calculation result and the optimal incident angle retrieval result.

[0024] Further, the Pearson correlation coefficient is:

[0025] ;

[0026] In the formula, indicates the mathematical expectation, cov indicates the covariance, A and B respectively indicate the sea surface backscattering data and the wind speed, and are respectively the mean and the standard deviation of A, and are respectively the mean and the standard deviation of B, A i and B irespectively represent the i-th observation value of variables A and B.

[0027] Further, the step S1 comprises:

[0028] (1) Obtain satellite observation data and buoy measurement data, and exclude invalid data from the satellite observation data and the buoy measurement data;

[0029] (2) Convert the wind speed data in the buoy measurement data into equivalent wind speed at 10m above the sea surface;

[0030] (3) Taking the buoy measurement data as the reference, set a time window and a space window, filter the satellite data falling into the window, and form a preliminary matching data pair;

[0031] (4) Quality identification screening is performed on the preliminary matching data pair, the matching data pair with the standard deviation of the sea surface backscattering data in the same buoy space-time window exceeding the set threshold is calculated and screened out, and the remaining matching data pair is averaged to generate a final matching data set.

[0032] Compared with the prior art, the present application has the following beneficial effects:

[0033] The present application effectively solves the problem of determining the optimal incident angle against disturbance in the electromagnetic scattering observation of the sea surface in the transition zone due to the dynamic coupling of the radar incident angle and the sea surface roughness. The traditional method is difficult to quickly adapt to changing sea conditions. The present application realizes the adaptive optimization selection of the incident angle for a specific sea condition for the first time by constructing a small slope approximation forward model and a wind field parameter inversion framework. The technical core is to generate wind speed disturbance samples and systematically evaluate the overall bias under different incident angles to accurately lock the incident angle with the strongest anti-disturbance ability under the current sea condition.

[0034] The present application significantly improves the stability of the scattering signal in the transition zone, so that the observation data remains low sensitivity to wind speed changes. This advantage directly translates into two application values: first, it greatly improves the accuracy and reliability of sea surface wind field remote sensing measurement, and provides better data basis for marine environment monitoring; second, by stabilizing the sea clutter characteristics, the detection ability of marine targets such as ships is effectively enhanced, which can significantly reduce the false alarm and missed detection probability, and has important practical significance for marine monitoring and target identification. BRIEF DESCRIPTION OF DRAWINGS

[0035] Figure 1 Flow chart of the transition zone anti-disturbance optimal incident angle retrieval method based on the small slope approximation method;

[0036] Figure 2 Flow chart for setting the initial guess wind speed and wind direction;

[0037] Figure 3The cost function changes with the wind speed when the real wind speed is 10 m / s;

[0038] Figure 4 The flowchart of searching for the optimal incident angle against disturbance;

[0039] Figure 5 The simulation results under the conditions of the reference wind speed and the disturbed wind speed when the inversion wind speed is 9.4 m / s;

[0040] Figure 6 The change trend of the weighted sum of squares of simulation differences with the incident angle when the inversion wind speed is 9.4 m / s;

[0041] Figure 7 The verification and evaluation of the searching result of the optimal incident angle. DETAILED DESCRIPTION

[0042] The application will be described in detail below with specific embodiments. The following examples will help those skilled in the art to further understand the application, but do not limit the application in any form. It should be noted that those skilled in the art can make several changes and improvements without departing from the concept of the application. These are within the scope of the application.

[0043] As shown in Figure 1 The implementation process of the transition zone optimal incident angle searching method against disturbance based on the small slope approximation method includes the following steps:

[0044] S1, obtaining satellite observation data and buoy measurement data, the satellite observation data including backscattering data and corresponding radar parameters of the marine area; the buoy measurement data including wind field information of the marine area; data preprocessing is performed on the satellite observation data and the buoy measurement data, and a matching data set for method verification is constructed.

[0045] First, data collection and preprocessing are performed. The experimental data used in the application includes satellite observation data and buoy measurement data. The satellite observation data includes backscattering data and corresponding radar parameters of the marine area, and the buoy measurement data includes wind speed, wind direction, latitude and longitude, and time. A matching data set for experimental verification is constructed by data preprocessing of the satellite observation data and the buoy measurement data.

[0046] Data preprocessing includes:

[0047] (1) obtaining satellite observation data and buoy measurement data, and excluding invalid data from the satellite observation data and the buoy measurement data.

[0048] Invalid data in satellite data and buoy data are identified and excluded through quality identification and filling data identification to ensure the validity of the data.

[0049] (2) Convert the wind speed data in the buoy measurement data into the equivalent wind speed at 10 m above sea level.

[0050] Currently, the commonly used data standard for sea surface wind speed inversion results, sea wave spectrum input, and geophysical model function input is the wind speed at 10 m above sea level. The distance from the sea surface to the anemometer installed on the buoy is not uniform, so it is necessary to convert the wind speed data measured by the buoy into the equivalent wind speed at 10 m above sea level.

[0051] ;

[0052] wherein, represents the wind speed at a height of 10 m above sea level.

[0053] (3) Set time window and space window based on buoy measurement data, filter satellite data falling into the window, and form a preliminary matching data pair.

[0054] To form a usable matching data set, the collected satellite data and auxiliary data need to be matched in time and space. The time window for time-space matching is set to 30 min, and the space window is set to 15 km. In time-space matching, the buoy data is taken as the reference, and satellite data within the time window and space window is filtered.

[0055] To form an experimental matching data set, the latitude and longitude and time information in the satellite data and buoy data need to be matched in time and space. The satellite data provides sea surface backscatter measured data and radar parameters (incident angle, azimuth) corresponding to the measured value, and the buoy data provides sea surface environmental information (wind speed, wind direction) corresponding to the measurement time. The time window for time-space matching is set to 30 min, and the space window is set to 15 km. In time-space matching, the buoy data is taken as the reference, and satellite data within the time window and space window is filtered to form a preliminary matching data.

[0056] (4) Quality identification screening of the preliminary matching data pair, calculation and screening of the matching data pair with sea surface backscatter data standard deviation exceeding the set threshold in the same buoy time-space window, and average processing of the remaining matching data pair to generate the final matching data set.

[0057] In order to improve the quality of the matching data set, the matching data pairs affected by measurement error need to be screened out. First, low-quality data is screened out through satellite data quality identification. On this basis, the standard deviation of all sea surface backscatter data in the same buoy data time-space window is calculated, and the part of the matching data with standard deviation exceeding 3 dB is screened out to reduce the influence of accidental error. Finally, multiple quality-controlled matching data pairs of the same buoy data time-space window are averaged to generate the matching data set.​

[0058] S2, based on the obtained backscattering data and corresponding radar parameters, using a small slope approximation model as a forward model, by constructing a cost function and iterative optimization, the reference wind speed and reference wind direction of the ocean area are obtained.

[0059] 2.1 Data preparation.

[0060] The data preparation section first needs to obtain sea surface backscattering observation data and corresponding radar parameters, and the radar parameters that need to be obtained include incident angle, azimuth angle, radar working waveband and polarization mode, etc. On this basis, according to the obtained data, the setting of the initial guess wind speed and the initial guess wind direction is carried out, and the input basis for the wind vector inversion work is established. The setting process of the initial guess wind speed and the initial guess wind direction is as shown in Figure 2 , including:

[0061] (1) The initial guess wind direction is fixedly set as the adverse wind direction, that is, the relative wind direction is 0°.

[0062] (2) The average value of the backscattering observation data of the sea surface in the same area is calculated, denoted as .

[0063] (3) The variables required for iteration are set: let .

[0064] (4) The radar parameters corresponding to the sea surface backscattering data are input into the small slope approximation model, and the sea surface backscattering when the wind speed is , and m / s is calculated, denoted as an array .

[0065] (5) The absolute deviation between and the array is calculated. If the absolute deviation is the smallest when the corresponding wind speed is , then let , return to (4) for iteration; if the absolute deviation is the smallest when the corresponding wind speed is , then let , return to (4) for iteration; if the absolute deviation is the smallest when the corresponding wind speed is , stop iteration.

[0066] (6) The initial guess wind speed is set to m / s.

[0067] 2.2 Cost function construction

[0068] After the data preparation is completed, the present application uses the maximum likelihood estimation method to perform wind vector inversion. First, a cost function required by the maximum likelihood estimation method is constructed, and the cost function is defined as a weighted sum of squares of differences between observation values and model simulation results:

[0069] ;

[0070] In the formula, represents the number of observation values, represents the variance of noise of the i th observation value, represents the i th observation value, and represent the wind speed and the relative wind direction, respectively, represents the backscattering simulation result. The value can be determined by the measurement accuracy of the observation data.

[0071] To ensure the statistical significance of the inversion result, the present application also performs dynamic cost threshold setting based on the chi-square distribution, and the significance level is set to 0.05. The cost threshold is determined according to the number of observation values and the inverse cumulative distribution function of the chi-square distribution.

[0072] 2.3 Obtaining the wind vector inversion result

[0073] The obtained radar parameters, initial guess wind speed and initial guess wind direction are input into the small slope approximation model to obtain the model simulation result under the initial guess condition. Then, the model simulation result and the observation data are input into the cost function to calculate the cost function value. The wind speed and the wind direction input into the sea surface electromagnetic scattering model are continuously adjusted to perform iterative operation. Taking the case where the true wind speed is 10 m / s as an example, the trend of the cost function with respect to the wind speed is shown in Figure 3 When the wind speed is 9.4 m / s, the cost function is minimum and does not exceed the cost threshold, so the iteration is stopped and the wind vector inversion result is output, and the reference wind speed and the reference wind direction of the marine area are obtained by inversion.

[0074] S3, a set of wind speed perturbation samples are generated in the neighborhood of the reference wind speed; the reference wind speed, the wind speed perturbation sample, the reference wind direction and the radar parameters are input into the small slope approximation model to retrieve the incident angle corresponding to the minimum total bias as the incident angle with the optimal anti-perturbation ability under the current sea state.

[0075] ​Specifically, on the basis of the wind vector inversion result obtained in step 2, the application uses a small slope approximation model to retrieve the optimal incidence angle under the corresponding sea surface roughness condition, the small slope approximation model belongs to a sea surface electromagnetic scattering model, and the sea surface electromagnetic scattering (including sea surface backscattering) is calculated by inputting radar parameters and a sea surface autocorrelation function, wherein the radar parameters to be input include an incident wave frequency, polarization, incidence angle and azimuth angle, and the autocorrelation function can be obtained by inverse Fourier transform of a sea wave spectrum under a specified wind speed and wind direction. The optimal incidence angle retrieval process is as shown in Figure 4 .

[0076] (1) The retrieval of the optimal incidence angle first takes the wind speed inversion result as a benchmark, generates a set of wind speed perturbation samples distributed at equal intervals in the neighborhood of the benchmark wind speed, and calculates the corresponding sea surface backscattering simulation data. The neighborhood range is set to ±1 m / s, and the generation interval is set to 0.1 m / s.

[0077] (2) The wind speed perturbation samples, the benchmark wind speed, the inversion wind direction and the radar parameters are input into the small slope approximation model to obtain the sea surface backscattering simulation results under different parameter conditions. When the inversion wind speed is 9.4 m / s, the simulation results under the conditions of the benchmark wind speed and the perturbed wind speed samples are as shown in Figure 5 , the simulation results under the conditions of the perturbed wind speed samples are concentrated near the simulation results under the condition of the benchmark wind speed, and with the increase of the incidence angle condition, the absolute deviation between the two types of simulation results first decreases and then increases, and there is an incidence angle with the minimum absolute deviation in the 10°-15° incidence angle interval. This shows that there is an incidence angle in the 10°-15° incidence angle interval, which has the lowest sensitivity to wind speed changes and the best ability to resist wind speed sudden disturbance.

[0078] (3) To balance the contribution difference of different perturbed wind speed samples, the weight coefficients are calculated according to the difference between each perturbed wind speed sample and the benchmark wind speed, and the weighted sum of squares of the deviation of the sea surface backscattering simulation data under the perturbed wind speed samples and the benchmark wind speed can be represented as follows.

[0079] ;

[0080] In the formula, represents the number of samples under the same incidence angle condition, represents the sea surface backscattering simulation result calculated by the sample wind speed, represents the sea surface backscattering simulation result calculated by the benchmark wind speed, The weight coefficient is represented. Considering that the distribution of the disturbance wind speed is usually uneven, and the disturbance wind speed close to the reference wind speed dominates, the weight coefficient is set to be inversely proportional to the wind speed difference between the sample wind speed and the reference wind speed, 1 is set at the reference wind speed, 0 is set at the maximum difference between the disturbance wind speed and the sample wind speed, and the intermediate part is determined by linear interpolation. The weight coefficient is inversely proportional to the wind speed difference between the sample wind speed and the reference wind speed, that is, the smaller the difference from the reference wind speed, the greater the weight.

[0081] (4) For different incidence angle conditions, the weighted sum of squares of the deviations between the simulation results of the sea surface backscattering under each disturbance wind speed sample and the reference wind speed is quantified, and is arranged according to the incidence angle and wind speed conditions. The weighted sum of squares of the simulation difference is used as an evaluation index. According to the change trend of the evaluation index with the incidence angle, the incidence angle with the strongest disturbance resistance, that is, the optimal incidence angle, is determined.

[0082] Taking the inversion wind speed as 9.4 m / s as an example, the change trend of the weighted sum of squares of the deviations with the incidence angle is shown in Figure 6 . The weighted sum of squares of the simulation difference decreases first and then increases with the increase of the incidence angle, and reaches the minimum value at the incidence angle of 11.5°, so 11.5° is determined as the optimal incidence angle under the current sea state condition.

[0083] S4, based on the matching data set constructed in step S1, the Pearson correlation coefficient between the backscattering data of the sea surface and the wind speed under different incidence angles is calculated, and the accuracy and effectiveness of the optimal incidence angle retrieval method are verified by comparing the zero point of the Pearson correlation coefficient with the optimal incidence angle obtained in step S3.

[0084] In order to verify and evaluate the accuracy and effectiveness of the optimal incidence angle retrieval method in the transition zone, the present application takes the wind speed of 10 m / s as an example, and uses the established matching data set to test the optimal incidence angle retrieval result, and the specific process is as follows.

[0085] (1) Data extraction. The sea surface backscattering data, radar parameters and wind field information are extracted from the obtained matching data set.

[0086] (2) Wind vector inversion. The wind vector inversion method in the present application is used to perform wind vector inversion to obtain the reference wind speed and the wind direction.

[0087] (3) Optimal incidence angle retrieval. The optimal incidence angle retrieval method in the present application is used to determine the optimal incidence angle.

[0088] (4) Pearson correlation coefficient calculation. The Pearson correlation coefficient between the sea surface backscattering data and the wind speed in the measured data is calculated under different incidence angle conditions. The calculation formula of the Pearson correlation coefficient can be represented as follows:

[0089] ;

[0090] wherein, and respectively represent sea surface backscattering data and wind speed, and respectively represent the mean and standard deviation of , and respectively represent the mean and standard deviation of , i and B i respectively represent the ith observation value of variable A and B.

[0091] (5) Taking the Pearson correlation coefficient as an evaluation index, the sensitivity of the sea surface backscattering data to the wind speed is quantified. When the Pearson correlation coefficient between the measured sea surface electromagnetic scattering data and the wind speed is zero, the sensitivity of the measured sea surface electromagnetic scattering data to the wind speed disturbance is the lowest, and the influence of the wind speed mutation is the smallest. Therefore, when the Pearson correlation coefficient between the measured sea surface electromagnetic scattering data and the wind speed is zero, the corresponding incident angle is the optimal anti-disturbance incident angle, that is, the zero point of the Pearson correlation coefficient calculation result is the optimal incident angle calculated by the measured data, and the difference between the zero point of the Pearson correlation coefficient calculation result and the optimal incident angle retrieval result is compared to realize the accuracy evaluation of the optimal incident angle retrieval result of the present application.

[0092] Figure 7 The optimal incident angle retrieval result and the Pearson correlation coefficient calculation result when the reference wind speed is 10 m / s are shown. The zero points of the Pearson correlation coefficient calculation result are almost completely consistent, indicating that the optimal incident angle retrieval method of the present application can effectively retrieve the optimal anti-disturbance incident angle.

[0093] The specific embodiments of the present application are described above. It needs to be understood that the present application is not limited to the above specific embodiments, and various changes or modifications can be made by those skilled in the art within the scope of the claims, which does not affect the essential content of the present application. In the case of no conflict, the embodiments of the present application and the features in the embodiments can be combined with each other arbitrarily.

Claims

1. A method for retrieving the optimal incident angle for disturbance resistance in the transition zone based on the small slope approximation method, characterized in that, include: S1. Acquire satellite observation data and buoy measurement data. Satellite observation data includes backscatter data of the ocean area and corresponding radar parameters; buoy measurement data includes wind field information of the ocean area. S1. Preprocess the satellite observation data and buoy measurement data to construct a matching dataset for method verification; S2. Based on the backscatter data and the corresponding radar parameters, use the small slope approximation model as the forward model, construct a cost function and perform iterative optimization to invert and obtain the reference wind speed and reference wind direction of the ocean area; S3. Generate a set of wind speed disturbance samples in the neighborhood of the reference wind speed; The reference wind speed, the wind speed disturbance sample, the reference wind direction, and the radar parameters are input into the small slope approximation model, and the incident angle corresponding to the minimum overall deviation is retrieved as the incident angle with the best anti-disturbance capability under the current sea state. S4. Based on the matching dataset constructed in step S1, calculate the Pearson correlation coefficient between the backscattered data of the sea surface and the wind speed under different incident angles. By comparing the zero point of the Pearson correlation coefficient with the optimal incident angle obtained in step S3, verify the accuracy and effectiveness of the optimal incident angle retrieval method.

2. The method for retrieving the optimal incident angle for transition zone disturbance resistance based on the small slope approximation method according to claim 1, characterized in that, In step S2, the wind vector inversion is performed using the maximum likelihood estimation method, and the cost function J is: ； In the formula, Indicates the number of observations. The variance of the noise for the i-th observation is represented by the following expression: Indicates the first One observation value, and These represent wind speed and relative wind direction, respectively. This represents the simulation results of backscattering.

3. The method for retrieving the optimal incident angle for transition zone disturbance resistance based on the small slope approximation method according to claim 1, characterized in that, Step S3 includes: (1) Based on the wind speed inversion results obtained by wind vector inversion, a set of equally spaced wind speed disturbance samples are generated in the neighborhood of the reference wind speed. (2) Input the wind speed disturbance sample, the reference wind speed, the inverted wind direction and the radar system parameters into the small slope approximation model, and calculate the reference wind speed and the sea surface backscattering simulation data corresponding to each wind speed disturbance sample respectively. (3) For each incident angle, calculate the weighted sum of squares of the deviations between the simulation results corresponding to each wind speed disturbance sample and the simulation results of the reference wind speed; (4) Based on the trend of the weighted sum of squares of the deviation changing with the incident angle, select the incident angle that minimizes the weighted sum of squares as the incident angle with the best resistance to environmental disturbance.

4. The method for retrieving the optimal incident angle for transition zone disturbance resistance based on the small slope approximation method according to claim 3, characterized in that, Weighting coefficients are calculated based on the differences between each wind speed disturbance sample and the reference wind speed, and the weighted sum of squared deviations of the sea surface backscattering simulation data at the wind speed samples and the reference wind speed are used. for: ； In the formula, This represents the number of samples under the same incident angle condition. This represents the simulation result of sea surface backscattering calculated using the wind speed of the j-th sample. This represents the simulation results of sea surface backscattering calculated using a reference wind speed. This represents the j-th weight coefficient.

5. The method for retrieving the optimal incident angle for transition zone disturbance resistance based on the small slope approximation method according to claim 2, characterized in that, Step S4 includes: (1) Extract sea surface backscattering data, radar parameters and wind field information from the obtained matching dataset; (2) Use the wind vector inversion method to perform wind vector inversion and obtain the reference wind speed and wind direction; (3) Determine the optimal angle of incidence using the optimal angle of incidence retrieval method; (4) Under different incident angles, calculate the Pearson correlation coefficient between sea surface backscattering data and wind speed in the measured data; (5) Using the Pearson correlation coefficient as an evaluation index, the sensitivity of sea surface backscattering data to wind speed is quantified, and the difference between the zero point of the Pearson correlation coefficient calculation result and the optimal incident angle retrieval result is compared.

6. The method for retrieving the optimal incident angle for transition zone disturbance resistance based on the small slope approximation method according to claim 5, characterized in that, Pearson correlation coefficient for: ； In the formula, Let represent the expected value, cov represent the covariance, and A and B represent the sea surface backscattering data and wind speed, respectively. and Let A be the mean and standard deviation, respectively. and The mean and standard deviation of B are respectively, and A is the mean and standard deviation of B. i and B i Let A and B represent the i-th observation values ​​of variables A and B, respectively.

7. The method for retrieving the optimal incident angle for transition zone disturbance resistance based on the small slope approximation method according to claim 1, characterized in that, Step S1 includes: (1) Acquire satellite observation data and buoy measurement data, and filter out invalid data from the satellite observation data and buoy measurement data; (2) Convert the wind speed data in the buoy measurement data into the equivalent wind speed at 10m above the sea surface; (3) Based on the buoy measurement data, set time windows and space windows, filter satellite data that fall within the windows, and form preliminary matching data pairs; (4) Quality identification screening is performed on the preliminary matching data pairs. Matching data pairs with standard deviation of sea surface backscatter data exceeding the set threshold within the same buoy spatiotemporal window are calculated and screened out. The remaining matching data pairs are averaged to generate the final matching dataset.

Images