A Microwave Radiometer Radiofrequency Interference Suppression Method Based on the Combination of Data Domain and Brightness Temperature Domain

Abstract

This invention discloses a microwave radiometer radiofrequency interference suppression method based on a combination of the data domain and brightness temperature domain. It primarily addresses the problem of degraded imaging performance caused by abnormal brightness temperature points during suppression in the original visibility data domain in existing technologies. The implementation scheme is as follows: Visibility data is obtained through complex correlation, and an optimized weight vector is calculated based on a selected optimization model; the visibility data and the optimized weight vector are weighted and summed to obtain a brightness temperature estimation image after one suppression step; peak points are found in the brightness temperature estimation image, and their status as brightness temperature anomalies is determined by an image threshold. If a peak point is found, its brightness temperature is suppressed until it falls below the brightness temperature anomaly threshold, and the brightness temperature estimation image is updated; this process of finding peak points is repeated until the entire brightness temperature estimation image falls below the image threshold, thus completing the suppression of microwave radiofrequency interference. This invention eliminates abnormal brightness temperature points near null regions, improves imaging quality, and can be used for integrated aperture passive microwave radiation measurements.

Description

Technical Field

[0001] This invention belongs to the field of microwave remote sensing technology, and specifically relates to a method for suppressing radio frequency interference in microwave radiometers, which can be used for integrated aperture passive microwave radiation measurement. Background Technology

[0002] Y-array passive remote sensing L-band interferometric integrated aperture radiometer systems are widely used in Earth and ocean observations due to their high resolution and wide field of view. Two typical applications are the microwave radiometer MIRAS carried by the SMOS satellite launched by the European Space Agency and the payload radiometer IMR carried by the China Ocean Salinity Satellite. MIRAS can be used to detect soil moisture and ocean salinity, while IMR can be used for ocean salinity detection. Traditional microwave radiometric measurements are susceptible to interference from non-natural signals in the same band, which can overwhelm the microwave radiation intensity of real objects. Due to the limitations of spatial frequency sampling, Gibbs tails are generated in high brightness temperature regions, which can disrupt or even hinder the inversion of geophysical parameters. Its impact has already been noted in the two Earth observation applications mentioned above, and this problem is expected to worsen. Based on the research experience of SMOS and the planning of future work, radio frequency interference (RFI) mitigation technology will be developed as an important part of the integrated aperture radiometer system design.

[0003] In passive remote sensing, radio frequency interference (RFI) signals are useless and harmful electromagnetic radiation that can damage receiver performance. Currently, the detection of RFI is increasing, and coupled with the widespread use of wireless technology globally, research institutions are increasingly concerned about RFI. Some RFI originates from illegal radiation within the protected frequency bands of passive observation, known as in-band RFI; others originate from lower-frequency harmonics emitted, known as out-of-band RFI. RFI has become a serious threat to passive remote sensing, especially in microwave radiometric measurements. Although microwave radiometers operate within protected frequency bands, their high sensitivity makes them highly susceptible to RFI. Therefore, airborne detection and mitigation techniques are needed to reduce the impact of RFI, improve the accuracy of radiometric measurements and the scientific rigor of data retrieval, and enhance spatial coverage in areas widely obscured by interference.

[0004] For over a decade, research institutions and experts in the field of radiometrics have been dedicated to researching methods to combat radio frequency interference (RFI). The most basic and effective method is to accurately locate the RFI and then have national authorities shut down illegal radio frequency interference sources in protected frequency bands. This approach has made some progress in the L-band. However, it is impossible to shut down all RFI sources, and new RFI sources will continuously emerge. Therefore, mitigating and eliminating the serious radio frequency interference problems faced in microwave radiometrics applications, and laying the foundation for integrated aperture microwave radiometrics in various application scenarios such as Earth remote sensing, is urgently needed.

[0005] The Xi'an Space Radio Technology Research Institute disclosed an "error correction method for a spaceborne synthetic aperture radiometer based on calibration field observation" in its patent application CN202211338438.1. The implementation steps are as follows: First, the spaceborne synthetic aperture radiometer is used to observe the target scene on Earth, obtain the relevant output measurement matrix of the target scene, and perform absolute calibration to obtain the calibrated target scene visibility function. Next, the spaceborne synthetic aperture radiometer is used to observe the calibration field, obtain the relevant output measurement matrix of the calibration field, and perform absolute calibration to obtain the calibrated calibration field visibility function. Then, the target scene visibility function and the calibration field visibility function are proportionally subtracted to obtain the difference visibility function. The difference visibility function is divided into Earth and space components. Finally, the brightness temperature is reconstructed using the Earth component difference visibility function, and summed with the sky brightness temperature of the difference model. The brightness temperature of the calibration scene model is proportionally compensated to the reconstructed difference brightness temperature to obtain the corrected reconstructed brightness temperature image. This method has two drawbacks. First, it requires visibility functions for both the target and calibration scenes, resulting in high computational cost and complexity. Second, it requires proportionally subtracting the visibility functions of the target and calibration scenes, and manually selecting this ratio can introduce new errors, exacerbate the trailing phenomenon, and worsen the reconstructed brightness and temperature image. Summary of the Invention

[0006] The purpose of this invention is to address the shortcomings of the prior art by proposing a microwave radiometer radiometer radiofrequency interference suppression method based on the combination of the data domain and the brightness temperature domain. This method reduces errors caused by human factors by suppressing the interference in the original visibility data domain, effectively mitigates the effects of radiofrequency interference (RFI) and its trailing effect, and improves the image quality of the reconstructed brightness temperature.

[0007] To achieve the above objectives, the technical solution of this aspect includes the following steps:

[0008] (1) Calculate the spatial frequency domain information of the scene by using pairwise correlated sparse array antennas, and use the calculation results as visibility data V;

[0009] (2) Select an optimization model based on application requirements, and solve the optimization model using the CVX quadratic programming mathematical software package to obtain the optimization weight vector ω corresponding to the inverted field of view. best ;

[0010] (3) The optimized weight vector and the visibility data are weighted and summed to obtain the brightness temperature estimation image of the inverted field of view;

[0011] (4) Find the peak point (ξ) of the brightness temperature estimation image. n ,η n Record its brightness temperature value as T. n And set a threshold based on the brightness temperature estimation image of the inverted field of view, and take the peak points that exceed the threshold as the brightness temperature abnormal points and the peak points that are less than the threshold as the brightness temperature normal points, where n represents the nth radio frequency interference point;

[0012] (5) Calculate the average brightness temperature T of N normal points around the abnormal brightness temperature point. average Array factor AF and initial step size factor γ;

[0013] (6) The brightness temperatures of outliers are suppressed based on the average brightness temperature, array factor, and step size factor to obtain the updated brightness temperatures T of the outliers. B and step size factor γ B ;

[0014] (7) Set the brightness temperature anomaly threshold to 1+r, and determine whether the brightness temperature of the current anomaly point needs to be further suppressed:

[0015] If T B / T average >1+r, return to step (6), where r is a user-defined constant;

[0016] Otherwise, update the brightness temperature estimation image and return to step (4);

[0017] (8) Repeat steps (4) to (7) until the brightness temperature of all abnormal points is lower than the threshold of the brightness temperature estimation image, and then end the suppression.

[0018] Compared with the prior art, the present invention has the following advantages:

[0019] First, the present invention obtains the first suppressed brightness temperature estimation image by directly weighting and summing the optimized weight vector obtained by the optimization model with the visibility data, avoiding further processing of the visibility data, and reducing errors caused by human factors and systematic errors caused by the visibility data processing process.

[0020] Secondly, this invention adds a brightness temperature estimation image threshold and an abnormal brightness temperature point threshold to the brightness temperature estimation image obtained by the first suppression. By setting the dual thresholds, abnormal brightness temperature points around the null are suppressed a second time, and the abnormal points are transformed into normal points. This overcomes the problem of abnormal brightness temperature points around the null caused by the first suppression in the prior art, and can effectively alleviate radio frequency interference and the resulting trailing phenomenon, thereby improving image quality. Attached Figure Description

[0021] Figure 1 This is a flowchart illustrating the implementation of the present invention;

[0022] Figure 2 This is a basic schematic diagram of the interferometric measurement principle in this invention;

[0023] Figure 3 This is an image showing the suppression of SMOS measured data numbered 199282533 using the present invention and existing technology;

[0024] Figure 4 This is an image showing the suppression of SMOS measured data numbered 199282415 using the present invention and existing technology;

[0025] Figure 5 This invention and existing technology are used to suppress radio frequency interference in simulated land scenes;

[0026] Figure 6 This invention and existing technology describe the suppression of radio frequency interference in a simulated land-sea boundary scenario. Detailed Implementation

[0027] The present invention will be further described in detail below with reference to the accompanying drawings and examples.

[0028] To enable those skilled in the art to better understand the present invention, the technical solutions of the present invention will be clearly and completely described below with reference to the accompanying drawings of the embodiments of the present invention. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, other embodiments obtained by those skilled in the art without creative effort should fall within the protection scope of the present invention.

[0029] It should be noted that the step numbers in the specification and claims of this invention are only for the purpose of clearly describing the embodiments of this invention and facilitating understanding, and their order is not limited.

[0030] Reference Figure 1 The implementation steps for this example are as follows:

[0031] Step 1: Calculate the raw visibility data.

[0032] refer to Figure 2 In this example, the scene plane is the plane where the target thermal radiation signal is located, S is the target thermal radiation signal, i and j are the two antennas used to receive the thermal radiation signal, the antenna plane is the plane where the two receiving antennas i and j are located, and the complex correlation receiver is the complex correlation operation performed after the two antennas receive the thermal radiation signal. Visibility data is obtained by performing complex correlation operations on the received signal. A brightness temperature estimation image is obtained by performing error correction and image inversion on the visibility data. The specific implementation is as follows:

[0033] Two receiving antennas i and j perform a complex correlation operation on the received target thermal radiation signal S. This involves sequentially performing co-directional and orthogonal correlation on the outputs of the two antennas. During orthogonal correlation, the received signal from one antenna is phase-shifted by π / 2 before being correlated with the other antenna. The results of these two correlations are the real and imaginary parts of the visibility data, respectively. Combining these two parts yields the visibility data V between the i-th and j-th antennas.

[0034]

[0035] Among them, T M (ξ,η) represents the scene brightness temperature distribution at the cosine of the (ξ,η) direction, u ij and v ij These are the baseline vectors corresponding to the i-th and j-th antennas, respectively.

[0036] Step 2: Solve for the optimal weight vector.

[0037] 2.1) Select an optimization model based on application requirements. Given the strong point source characteristics of RF interference, a null region needs to be defined in the model, dividing the entire field of view into the null region and the desired direction. This example follows the principle that the brightness temperature gain remains constant in the desired direction, while the brightness temperature gain in the null region is equal to the null depth parameter, determining the following optimization model:

[0038]

[0039] st

[0040]

[0041]

[0042] Where, ||ω best ||2 represents the optimization weight vector ω best The binary norm of , (·) T Indicates the transpose operation, (ξ) k ,η k f(ξ) represents the position of the brightness temperature point outside the null region in the inversion field of view.k ,η k ) is the signal steering vector matrix, (ξ) i ,η i f(ξ) represents the location of the brightness temperature point in the zero-depression region. i ,η i U is the signal steering vector matrix. N ε represents the required zero-depression region, where ε is the corresponding zero-depression depth.

[0043] 2.2) The optimization model was solved using the CVX quadratic programming software package to obtain the optimization weight vector ω corresponding to the inverted field of view. best .

[0044] Step 3: Obtain the brightness temperature estimation image of the inverted field of view and identify brightness temperature anomalies.

[0045] 3.1) Optimize the weight vector ω best The brightness temperature estimate of the inverted field of view is obtained by weighted summation with the visibility data V, after first suppression:

[0046]

[0047] Where T is the inverted brightness temperature of the brightness temperature estimation image, and ω best To optimize the weight vector, (·) T This indicates a transpose operation, where V represents the visibility data.

[0048] 3.2) Find the peak point (ξ) of the brightness temperature estimation image. n ,η n Record its brightness temperature value as T. n , where n represents the nth radio frequency interference point;

[0049] 3.3) Set a threshold based on the brightness temperature estimation image of the inverted field of view, that is, use the MATLAB built-in function mean(T) to calculate the total brightness temperature mean of the brightness temperature estimation image, and use it as the threshold P of the brightness temperature image;

[0050] 3.4) The peak brightness temperature T of the brightness temperature estimation image. n Compare with the image threshold P:

[0051] If T n If the value is greater than P, then the peak value is considered an abnormal brightness temperature point, and step 4 is executed.

[0052] If T n If the value is less than or equal to P, then the peak value is considered a normal brightness temperature point.

[0053] Step 4: Calculate the parameters of the brightness temperature anomaly point.

[0054] 4.1) Calculate the average brightness temperature T of N normal points surrounding the abnormal brightness temperature point.average :

[0055]

[0056] Where T i Let N be the brightness temperature value of the i-th normal point around the abnormal brightness temperature point. The brightness temperature values ​​of these normal points can be obtained directly from the inverted brightness temperature T of the brightness temperature estimation image. N is the number of normal points around the abnormal brightness temperature point. In this example, N is a positive integer ≤ 6.

[0057] 4.2) Calculate the array factor AF:

[0058]

[0059] Where Δs = Δu·Δv, Δu and Δv are the minimum sampling intervals along the ξ-axis and η-axis, respectively, and u and v are the baseline vectors corresponding to the i-th and j-th antennas, respectively. n ,η n () represents the coordinates of the brightness temperature anomaly point;

[0060] 4.3) Calculate the initial step size factor γ:

[0061]

[0062] Where T is the brightness temperature value of the brightness temperature estimated image, and mean(T) is the average brightness temperature of the total brightness temperature estimated image. n This represents the brightness temperature value at the point of brightness temperature anomaly.

[0063] Step 5: Perform secondary suppression on the brightness temperature of abnormal brightness points.

[0064] 5.1) Calculate the updated abnormal point brightness temperature T B :

[0065] Based on the parameters of the abnormal brightness temperature point, the array factor, the initial step size factor, and the brightness temperature of the abnormal point are multiplied together. Then, the result of the multiplication is subtracted from the brightness temperature of the abnormal brightness temperature point to obtain the updated brightness temperature T of the abnormal point. B This allows for an update of brightness temperature anomalies:

[0066] T B =T n -AF·γ·T n

[0067] Where T n is the brightness temperature value of the brightness temperature anomaly point, AF is the array factor, and γ is the initial step size factor.

[0068] 5.2) Calculate the updated step size factor γ based on the brightness temperature values ​​of the brightness temperature anomalies. B :

[0069]

[0070] Among them, T overage The average brightness temperature of N normal points surrounding the abnormal brightness temperature point.

[0071] 5.3) Set the brightness temperature anomaly threshold to 1+r, and determine whether the brightness temperature of the current anomaly point needs to be further suppressed:

[0072] If T B / T average If the value is greater than 1+r, then return to step 5.1) and continue to update the brightness temperature anomaly point until the requirement is not met, and complete the suppression of the current abnormal brightness temperature point, where r is a user-defined constant;

[0073] Otherwise, update the brightness temperature estimation image and return to step 3;

[0074] 5.4) Repeat steps 3 to 5.3) until the brightness temperature of all abnormal points is lower than the threshold of the brightness temperature estimation image, then end the suppression and complete the secondary suppression of the entire brightness temperature estimation image.

[0075] The effects of this invention will be further illustrated below with simulation experiments:

[0076] 1. Simulation experimental conditions:

[0077] The software platform for the simulation experiment of this invention is: Windows 11 operating system, MATLAB R2021a and CVX quadratic programming mathematical toolkit.

[0078] The SMOS visibility data used in the dataset is L1a level data of soil moisture and ocean salinity from the SMOS satellite downloaded from the ESA website. The format is DBL file. After downloading, two sets of visibility data were selected and exported in txt format.

[0079] The measured data used in the simulation experiment were the measured data with snapshot numbers 199282533 and 199282415 respectively.

[0080] The simulation data used in the simulation experiment consisted of visibility data from a land scene with an average brightness temperature of 290K and an RFI source with an RFI source of 2000K added, and visibility data from a land-sea boundary scene with an RFI source of 2000K added. The brightness temperature of the part where ξ < 0.3 was set to 290K, representing the land scene, and the brightness temperature of the part where ξ > = 0.3 was set to 120K, representing the ocean scene.

[0081] 2. Simulation content and result analysis.

[0082] Simulation Experiment 1: Radio frequency interference (RFI) suppression was performed on the measured data of snapshot number 199282533 using both the present invention and existing array factor synthesis methods. The results are as follows... Figure 3 As shown, the horizontal axis in the image represents azimuth ξ, and the vertical axis represents azimuth η. The hexagon in the image is the brightness temperature estimation image. The darker areas in the image represent the Earth's background brightness temperature, and the point sources brighter than the background brightness temperature are RFI point sources. The brighter areas in the six directions near the point sources are the trailing caused by RFI. The long color plot on the right side of the image is a color scale filling the hexagonal field of view, where:

[0083] Figure 3 (a) is the inversion image of the SMOS measured data with the snapshot number 199282533. The brightness temperature range is [0-350], and the unit is K. It can be clearly seen from the figure that RFI and its tail seriously degrade the imaging performance of the inversion image.

[0084] Figure 3 (b) is the result of RFI mitigation of the measured data in Experiment 1 using the array factor synthesis algorithm that minimizes the existing weight vector norm. The brightness temperature range is [0-350], and the unit is K.

[0085] Figure 3 (c) The result of RFI mitigation of the measured data in Experiment 1 using the method of the present invention, with a brightness temperature range of [0-350] and unit K;

[0086] Figure 3 (d) is Figure 3 (b) Subtract Figure 3 (c) is a brightness temperature distribution image, which shows the advantages of the method of the present invention compared with the prior art. The brightness temperature range of the image is [0-100], and the unit is K.

[0087] Simulation Experiment 2: Radio frequency interference (RFI) suppression was performed on the measured data of snapshot number 199282415 using both the present invention and existing array factor synthesis methods. The results are as follows: Figure 4 As shown, the horizontal axis in the image represents azimuth ξ, and the vertical axis represents azimuth η. The hexagon in the image is the brightness temperature estimation image. The darker areas in the image represent the Earth's background brightness temperature, and the point sources brighter than the background brightness temperature are RFI point sources. The brighter areas in the six directions near the point sources are the trailing caused by RFI. The long color plot on the right side of the image is a color scale filling the hexagonal field of view, where:

[0088] Figure 4 (a) is the inversion image of the SMOS measured data with the snapshot number 199282415. The brightness temperature range is [0-350] and the unit is K. It can be clearly seen from the figure that RFI and its tail seriously degrade the imaging performance of the inversion image.

[0089] Figure 4 (b) is the result of RFI mitigation of the measured data in Experiment 2 using the array factor synthesis algorithm that minimizes the existing weight vector norm. The brightness temperature range is [0-350], and the unit is K.

[0090] Figure 4 (c) The result of RFI mitigation of the measured data in Experiment 2 using the method of the present invention, with a brightness temperature range of [0-350] and unit K;

[0091] Figure 4 (d) is Figure 4 (b) Subtract Figure 4 (c) is a brightness temperature distribution image, which shows the advantages of the method of the present invention compared with the prior art. The brightness temperature range of the image is [0-100], and the unit is K.

[0092] from Figure 3 and Figure 4 As can be seen, the present invention can effectively remove abnormal brightness temperature points existing in the prior art, thereby improving the quality of brightness temperature estimation images.

[0093] Simulation Experiment 3: Using both the present invention and existing array factor synthesis methods, radio frequency interference suppression was performed on visibility data of a simulated land scene with an average brightness temperature of 290K and an added 2000K RFI source. The results are as follows: Figure 5 As shown, the horizontal axis in the image represents azimuth ξ, and the vertical axis represents azimuth η. The hexagon in the image is the brightness temperature estimation image. The darker areas in the image represent the Earth's background brightness temperature, and the point sources brighter than the background brightness temperature are RFI point sources. The brighter areas in the six directions near the point sources are the trailing caused by RFI. The long color plot on the right side of the image is a color scale filling the hexagonal field of view, where:

[0094] Figure 5 (a) is the brightness temperature distribution after adding a 2000K RFI source to a simulated land scene with an average brightness temperature of 290K. The brightness temperature range is [0-350], and the unit is K. It can be clearly seen from the figure that the RFI and its trailing seriously degrade the imaging performance of the inverted image.

[0095] Figure 5 (b) is the result of RFI mitigation of the simulated data in Experiment 3 using the array factor synthesis algorithm that minimizes the existing weight vector norm. The brightness temperature range is [0-350], and the unit is K.

[0096] Figure 5 (c) is the result of RFI mitigation of the simulated data in Experiment 3 using the method of the present invention. The brightness temperature range is [0-350], and the unit is K.

[0097] Figure 5 (d) is Figure 5 (b) Subtract Figure 5 (c) is a brightness temperature distribution image, which shows the advantages of the method of the present invention compared with the prior art. The brightness temperature range of the image is [0-100], and the unit is K.

[0098] Simulation Experiment 4 uses both the present invention and existing array factor synthesis methods to suppress radio frequency interference (RFI) in a simulated land-sea interface scenario with 2000K RFI source visibility data. The results are as follows: Figure 6 As shown, the horizontal axis in the image represents azimuth ξ, and the vertical axis represents azimuth η. The hexagon in the image is the brightness temperature estimation image. The darker areas in the image represent the Earth's background brightness temperature, and the point sources brighter than the background brightness temperature are RFI point sources. The brighter areas in the six directions near the point sources are the trailing caused by RFI. The long color plot on the right side of the image is a color scale filling the hexagonal field of view, where:

[0099] Figure 6 (a) is the brightness temperature distribution map after adding a 2000K RFI source to the simulated land-sea boundary scene. The brightness temperature range is [0-350], and the unit is K. It can be clearly seen from the figure that the RFI and its tail seriously degrade the imaging performance of the inverted image.

[0100] Figure 6 (b) is the result of RFI mitigation of the simulated data in Experiment 4 using the array factor synthesis algorithm that minimizes the existing weight vector norm. The brightness temperature range is [0-350], and the unit is K.

[0101] Figure 6 (c) The result of using the method of the present invention to simulate the data in Experiment 4 after RFI mitigation, with a brightness temperature range of [0-350] and unit K;

[0102] Figure 6 (d) is Figure 6 (b) Subtract Figure 6 (c) is a brightness temperature distribution image, which shows the advantages of the method of the present invention compared with the prior art. The brightness temperature range of the image is [0-100], and the unit is K.

[0103] from Figure 5 and Figure 6 As can be seen, the present invention can overcome the problem of abnormal bright temperature points around zero depression caused by the use of existing technology, effectively alleviate radio frequency interference and the resulting trailing phenomenon, and improve image quality.

[0104] The array factor synthesis algorithm using the existing weight vector norm minimization method described above in the simulation experiment is a visibility data domain suppression method based on minimizing measurement noise variance, first proposed by Jun Li, Fei Hu, Feng He, Liang Wu, and Xiaohui Peng in their paper "SMOS RFI Mitigation Using Array Factor Synthesis of Synthetic Aperture Interferometric Radiometry." (IEEE International Geoscience and Remote Sensing Symposium (IGARSS), 2016, 820-823). The specific steps of this method are: First, calculate the visibility function, i.e., the output complex correlation of the sparse array antenna pairs; second, select an optimization model according to the actual application requirements and solve for the corresponding optimized weight vector; third, sum the weight vector with the visibility function to obtain the brightness temperature estimation image of the observed scene.

[0105] 3. Perform quantitative analysis on simulation experiments 3 and 4 to evaluate the RFI mitigation performance of the present invention.

[0106] Using the interference-free Fourier imaging results under ideal conditions as the comparison standard, the root mean square of the difference between the RFI mitigation results under interference conditions and the standard is calculated as the interference residual RMSE:

[0107]

[0108] Among them, T i 'T' is the brightness temperature of the i-th pixel in a Fourier image without added interference, used as a standard for comparing interference suppression performance. i Let N be the brightness temperature of the i-th pixel in the brightness temperature estimation image corresponding to different suppression methods after adding radio frequency interference, and N be the number of discrete pixels in the inverted image.

[0109] Tables 1 and 2 show the residual error RMSE calculated using the array factor synthesis algorithm and the method of this invention for simulation experiments 3 and 4, respectively.

[0110] Simulation Experiment 3 shows the residual errors obtained by the two methods in a land scenario, as shown in Table 1.

[0111] Simulation Experiment 4 shows the residual errors obtained by the two methods in a land-sea interface scenario, as shown in Table 2.

[0112] Table 1 Land Scene

[0113]

[0114] Table 2 Scenarios at the Land-Sea Boundary

[0115]

[0116] As can be seen from Tables 1 and 2, the interference residuals using the method of the present invention are lower, the suppression effect is better, and the imaging performance can be better optimized.

[0117] The simulation experiments above show that the method of this invention uses a radio frequency interference source suppression method based on the combination of the data domain and the brightness temperature domain. It suppresses the radio frequency interference source in both the visibility data domain and the brightness temperature domain. The background brightness temperature is restored by using the quadratic programming mathematical toolkit CVX in conjunction with MATLAB programming. This effectively alleviates the adverse effects caused by RFI and its trailing effect, solves some abnormal brightness temperature points near the null region caused by excessive suppression in existing technologies, and optimizes imaging performance. It is a very practical radio frequency interference mitigation method.

Claims

1. A method for suppressing radio frequency interference in microwave radiometers based on a combination of data domain and brightness temperature domain, characterized in that, Including the following: (1) Calculate the spatial frequency domain information of the scene by using pairwise correlated sparse array antennas, and use the calculation results as visibility data V; (2) Select an optimization model based on application requirements, and solve the optimization model using the CVX quadratic programming mathematical software package to obtain the optimization weight vector ω corresponding to the inverted field of view. best ; (3) The optimized weight vector and the visibility data are weighted and summed to obtain the brightness temperature estimation image of the inverted field of view; (4) Find the peak point (ξ) of the brightness temperature estimation image. n ,η n Record its brightness temperature value as T. n And set a threshold based on the brightness temperature estimation image of the inverted field of view, and take the peak points that exceed the threshold as the brightness temperature abnormal points and the peak points that are less than the threshold as the brightness temperature normal points, where n represents the nth radio frequency interference point; (5) Calculate the average brightness temperature T of N normal points around the abnormal brightness temperature point. average Array factor AF and initial step size factor γ; (6) The brightness temperatures of outliers are suppressed based on the average brightness temperature, array factor, and step size factor to obtain the updated brightness temperatures T of the outliers. B and step size factor γ B ; (7) Set the brightness temperature anomaly threshold to 1+r, and determine whether the brightness temperature of the current anomaly point needs to be further suppressed: If T B / T average >1+r, return to step (6), where r is a user-defined constant; Otherwise, update the brightness temperature estimation image and return to step (4); (8) Repeat steps (4) to (7) until the brightness temperature of all abnormal points is lower than the threshold of the brightness temperature estimation image, and then end the suppression.

2. The method according to claim 1, characterized in that, In step (1), the spatial frequency domain information of the scene is calculated using pairwise correlated sparse array antennas, as shown in the following formula; Where V represents the visibility data between the i-th antenna and the j-th antenna, and T... M (ξ,η) represents the scene brightness temperature distribution at the cosine of the (ξ,η) direction, u ij and v ij These are the baseline vectors corresponding to the i-th and j-th antennas, respectively.

3. The method according to claim 1, characterized in that, The optimization model selected in step (2) is represented as follows: st Where, ||ω best ||2 represents the optimization weight vector ω best The binary norm of , (·) T Indicates the transpose operation, (ξ) k ,η k f(ξ) represents the position of the brightness temperature point outside the null region in the inversion field of view. k ,η k ) is the signal steering vector matrix, (ξ) i ,η i f(ξ) represents the location of the brightness temperature point in the zero-depression region. i ,η i U is the signal steering vector matrix. N ε represents the desired zero-depression region, and ε is the corresponding zero-depression depth.

4. The method according to claim 1, characterized in that, The brightness temperature estimation image of the inverted field of view obtained in step (3) is described as follows: Where T is the inverted brightness temperature of the brightness temperature estimation image, and ω best To optimize the weight vector, (·) T This indicates a transpose operation, where V represents the visibility data.

5. The method according to claim 1, characterized in that, In step (5), the average brightness temperature T is calculated. average The formula is as follows: Where T i Let N be the brightness temperature values ​​of N normal points surrounding the abnormal brightness temperature point, where N is a positive integer ≤ 6.

6. The method according to claim 1, characterized in that, In step (5), the array factor AF is calculated using the following formula: Where Δs = Δu·Δv, Δu and Δv are the minimum sampling intervals along the ξ-axis and η-axis, respectively, and u and v are the baseline vectors corresponding to the i-th and j-th antennas, respectively. n ,η n () represents the coordinates of the brightness temperature anomaly point.

7. The method according to claim 1, characterized in that, In step (5), the initial step size factor γ is calculated using the following formula: Where T is the brightness temperature value of the brightness temperature estimated image, and mean(T) is the average brightness temperature of the total brightness temperature estimated image. n This represents the brightness temperature value at the point of brightness temperature anomaly.

8. The method according to claim 1, characterized in that, The updated abnormal illumination temperature T is obtained in step (6). B , means as follows: T B =T n -AF·γ·T n Where T n is the brightness temperature value of the brightness temperature anomaly point, AF is the array factor, and γ is the initial step size factor.

9. The method according to claim 1, characterized in that, The updated step size factor γ is obtained in step (6). B , means as follows: Among them, T average T is the average brightness temperature of N normal points surrounding the abnormal brightness temperature point. n This represents the brightness temperature value at the point of brightness temperature anomaly.

Images