A polarization calibration method for polarization GNSS-R receiver

Abstract

The present invention provides a polarization calibration method for a polarization GNSS-R receiver, including: obtaining a delay-Doppler map DDM; obtaining Stokes parameters from the delay-Doppler map DDM; defining one or more non-ideal effects that need to be compensated; calibrating the non-ideal effects based on the Stokes parameters; converting the factors that need to be compensated for different non-ideal effects into error matrices of the same dimension; calibrating the Stokes parameters based on the error matrix; and inverting and restoring the calibrated Stokes parameters into a delay-Doppler map DDM map. The present invention solves the gain loss problem caused by the difference between the polarization direction of the receiver antenna and the polarization direction of the reflected signal, the phase shift problem caused by the bidirectional Faraday rotation of the reflected signal after passing through the ionosphere twice, and the non-ideal state problems such as elliptical and phase shift caused by the non-ideal axial ratio of the GNSS transmitting antenna.

Description

Technical Field

[0001] The present invention relates to the technical field of global navigation satellite systems (GNSS-R), and in particular to a polarization calibration method for a polarization GNSS-R receiver. Background Art

[0002] GNSS-R technology is a new technology in the field of remote sensing. It can be used to study climate change, ocean roughness and salinity, ice, wind speed, disaster monitoring, atmospheric and ionosphere measurements, etc. In terms of soil detection, polarimetric GNSS-R has been proven to be sensitive to different land parameters, such as freeze-thaw, crop growth or soil moisture. The GNSS-R satellites currently in orbit include but are not limited to UK-TDS-1, Cyclone GNSS (CYGNSS), Fengyun-1, Fengyun-3E and F, and SMAP-R using passive radiometers. All of these dedicated missions carry a GNSS-R receiver as the main instrument and have successfully collected global navigation satellite system reflection signals in the microwave band.

[0003] In addition to the physical information of the rough surface, the received satellite-borne GNSS reflected signals also contain errors from factors such as the ionosphere and troposphere and the non-ideal nature of the receiver payload itself, which will seriously affect the quality of subsequent products, especially the accuracy of some parameters obtained based on the inversion of the two-dimensional delay Doppler map (DDM), such as sea breeze, sea ice, soil moisture, vegetation moisture content, etc. Therefore, the GNSS-R receiver payload will perform on-site calibration on the relevant product data.

[0004] In GNSS-R, in situ calibration focuses on an observable: the bistatic radar cross section (BRCS). This observable is usually calibrated using the assumption that the transmitted signal is perfectly circularly polarized, and some known parameters such as the receiver antenna pattern gain or the receiver noise floor are used. In addition, the power transmitted by each GNSS satellite is required, which is usually estimated by measuring the direct signal with the same receiver or using a lookup table and information from the carrier-to-noise ratio (C / N0) value of the system GNSS receiver.

[0005] The above payload field calibration scheme is sufficient for data currently used only in the ocean, because in marine applications, most reflected signals are circular and the reflected signals are left-hand circularly polarized (LHCP). However, in future missions, the payload of polarized GNSS-R will receive elliptical reflected signals from complex surfaces such as soil, ice, and atmosphere, so appropriate calibration schemes are needed to deal with non-ideal conditions such as elliptical and phase shift of the transmitted signal. At the same time, since the polarization plane of the electromagnetic wave rotates continuously with the forward direction as the axis in the magnetized plasma when the linear polarized wave propagates along the magnetic field direction in the magnetized plasma, the signal will experience the rotation effect caused by the ionosphere on the way to the surface reflection point and on the way to the receiver, resulting in the misalignment of the signal polarization direction with the polarization direction of the receiving antenna, causing the attenuation of the received power. In addition, there is also the influence of the misalignment of the polarization plane (X / Y plane) of the receiving antenna itself and the polarization plane (H / V plane) of the scattering surface. These defects seriously affect the quality of GNSS-R products and must be compensated accordingly.

[0006] For example, in the prior art, an invention patent with publication number CN115508867A discloses a GNSS-R receiver dual-antenna signal collaborative correlation processing system and method, which uses the GNSS direct signal received by the upward-looking antenna to estimate the signal's carrier-to-noise ratio, satellite Doppler information, and specific parameters such as code phase, and uses this information to reconstruct the correlation peak of the direct signal leakage interference, and generate a corresponding DDM image to assist the correlation process between the reflected signal received by the downward-looking antenna and the local code, and remove and reconstruct the DDM in the reflected correlation waveform, thereby suppressing the direct leakage interference and ultimately improving the quality of the output DDM. It uses the method of removing and reconstructing the DDM of the reflected correlation waveform for calibration. Summary of the invention

[0007] A brief summary of embodiments of the present invention is provided below to provide a basic understanding of certain aspects of the present invention. It should be understood that the following summary is not an exhaustive summary of the present invention. It is not intended to identify key or important parts of the present invention, nor is it intended to limit the scope of the present invention. Its purpose is merely to provide certain concepts in a simplified form as a prelude to a more detailed description discussed later.

[0008] In order to solve the non-ideal factors of the GNSS-R receiver from the signal and the load, the idea of ​​the present invention is to perform polarization calibration on the delay Doppler map DDM, and then use the processed delay Doppler map DDM to replace the I / Q polarization data. By comparing the I / Q data stream and generating DDM, the satellite-processed delay Doppler map DDM is used as the calibration input, which has a lower data rate and a higher signal-to-noise ratio, making it easier to perform polarization calibration.

[0009] Specifically, a polarization calibration method for a polarization GNSS-R receiver of the present invention includes:

[0010] Step 1: Obtain a Delay-Doppler Mapping DDM (Delay-Doppler Mapping); obtain Stokes parameters from the Delay-Doppler Map DDM;

[0011] Step 2: Define one or more non-ideal effects that need to be compensated.

[0012] Step 3: Calibrate the non-ideal effects defined in step 2 based on the Stokes parameters;

[0013] Step 4: Convert the factors that need to be compensated for the different non-ideal effects in step 3 into error matrices of the same dimension;

[0014] Step 5: Calibrate the Stokes parameters;

[0015] Step 6: Invert and restore the calibrated Stokes parameters into the DDM diagram, thereby achieving calibration of the DDM diagram.

[0016] Furthermore, in step 1, obtaining Stokes parameters from the delay Doppler map DDM specifically includes: constructing a polarization scenario, extracting the polarization characteristics of the GNSS reflection signal, and obtaining Stokes parameters; the Stokes parameters are parameters that describe the polarization state of electromagnetic waves, and the formula for extracting Stokes parameters is:

[0017] S0=<|E RH | 2 >+<|E RV | 2 >;

[0018] S1=<|E RH | 2 >-<|E RV | 2 >;

[0019]

[0020]

[0021] Where: E RH and E RV It is the complex electric field after the two linear orthogonal polarization components (H and V) of the right-hand circularly polarized transmitted signal are cross-correlated with the C / A code replica. Stokes parameters S0-S3 are often used to study the polarization state of light or electromagnetic waves. They are prior art and are described as follows:

[0022] S0 (intensity parameter) represents the total electric field strength, which is usually expressed in units of 1, and the other three values ​​are normalized accordingly;

[0023] S1 (polarization parameter) represents the linear polarization degree of electromagnetic waves, ranging from -1≤S1≤1;

[0024] S2 (polarization parameter) represents the linear polarization degree of electromagnetic waves. Unlike S1, it measures the polarization degree perpendicular to S1, with a range of -1≤S2≤1.

[0025] S3 (circular polarization parameter) represents the degree of circular polarization, which measures the rotational properties of electromagnetic waves, and has a range of -1≤S3≤1.

[0026] As a feasible solution, the non-ideal effect in step 2 includes polarization calibration of the receiver; and the calibration of the non-ideal effect defined in step 2 based on the Stokes parameter in step 3 specifically includes:

[0027] Calculate the direction in which the axis of the receiver antenna is pointing:

[0028] δθ=θ sp -θ T ;

[0029]

[0030] where θ T and are the azimuth and elevation angles of the actual pointing direction of the receiver antenna, θ sp and are the azimuth and elevation angles at the mirror reflection point, respectively. The angle calculation uses the Earth-centered Earth-fixed coordinate system. δθ and It is the azimuth and elevation angle offset in the clockwise direction with the receiver position as the origin and the mirror reflection point as the target. The angle calculation uses the Earth-centered Earth-fixed coordinate system.

[0031] As a feasible solution, the non-ideal effect in step 2 includes two Faraday rotation (FAR) calibrations on the reflection path; the calibration of the non-ideal effect defined in step 2 based on the Stokes parameters in step 3 specifically includes:

[0032] Estimate the Faraday rotation angle Ω, the estimation method is as follows:

[0033] Ω≈λ 2 TEC U ;

[0034] Where λ is the wavelength of the reflected signal, TEC U is the normalized total column electron content, obtained by inverting the TEC map provided by GPS.

[0035] As a feasible solution, the non-ideal effect in step 2 includes calibration of the transmitter signal axial ratio; and the calibration of the non-ideal effect defined in step 2 based on the Stokes parameters in step 3 specifically includes:

[0036] The full polarization model of the reflected signal is inverted using GNSS-R L1 level data (GNSS-R L1 level data observed by GNSS satellites) and the ERA5 dataset;

[0037] Performing preliminary calibration on the Stokes parameters obtained in step 1, including Faraday rotation calibration and receiver polarization calibration;

[0038] Set up the optimizer and loss function to obtain the calibrated Stokes parameters.

[0039] Among them, a feasible solution is: the minimized loss function is as follows:

[0040]

[0041] Where m is the antenna axial ratio, S n (m)(n=0,1,2,3) is the full polarization model with unknown parameter m, is the corrected Stokes parameter.

[0042] In the above minimized loss function, the label is The training set is S n (m), S n (m) = S n +me jφ , S n is the Stokes parameter when the antenna polarization calibration is not performed. It is a constant and is fitted by setting the above formula and label. m is the antenna axial ratio, which is an unknown number and is required when calibrating the non-ideal effects of the transmitter antenna.

[0043] Furthermore, the step 4 converts the factors that need to be compensated for the different non-ideal effects in step 3 into error matrices of the same dimension, specifically including:

[0044] Define a matrix A to describe the error factors of the receiving antenna and the transmitting antenna:

[0045]

[0046] Where m is the antenna axial ratio, is the phase difference between H and V polarizations; σ1 and σ2 are the normalized powers of H and V polarizations, respectively, e jφ =cos(φ)+jsin(φ);

[0047] m is the antenna axial ratio, the receiver antenna (A x represents the amplitude in the x-polarization direction, A y represents the amplitude in the y-polarization direction):

[0048]

[0049] e jφ middle is the phase difference (angle) between x and y polarizations. jφ =cos(φ)+jsin(φ)(Euler formula), when the phase deviation angle is equal to 90 degrees (no error condition), e jφ =1, then only the influence of the antenna axial ratio remains, m=1 is circular polarization, m≠1 is positive elliptical polarization. When the phase deviation angle is not 90 degrees, the antenna is partially elliptical polarization.

[0050] Define the matrix C as the error factor describing the Faraday rotation:

[0051]

[0052] Where Ω is the Faraday rotation angle obtained in step 3;

[0053] Furthermore, in step 5, the calibration of the Stokes parameters is as follows:

[0054]

[0055] In the formula and S n are the calibrated and uncalibrated Stokes parameters, n=0, 1, 2, 3; and are the calibration matrices for the transmitter and receiver, respectively, C -1 is the calibration matrix for the Faraday rotation.

[0056] Wherein, the step 1: obtaining the delay Doppler map DDM specifically includes:

[0057] Step 11: Generate a GNSS-R data set, which includes a data set received by a direct antenna and a data set received by a reflection antenna. The data set received by the direct antenna includes: an I / Q data set received by the direct antenna, GNSS satellite PVT (Position, Velocity and Time) information, and satellite clock error; the data set received by the reflection antenna includes an I / Q data set of H and V channels received by the reflection antenna. The receiver uses two orthogonal linear polarization receiving antennas, and can use a pair of circularly polarized antennas with opposite meanings to receive vertically polarized, horizontally polarized, and cross-polarized reflected signals.

[0058] Step 12: Calculate the mirror reflection point position and reflection timestamp based on the GNSS-R dataset. Specifically, the mirror reflection point position (SP) is calculated by finding a following function that minimizes the reflection path length and satisfies the condition that the incident angle at the mirror reflection point position is equal to the reflection angle. Then, the mirror reflection point position is constrained to the earth's surface. The following function is as follows:

[0059]

[0060] In the formula are the position vectors of the GNSS-R receiver, the mirror reflection point, and the GNSS satellite in the Earth-centered Earth-fixed rectangular coordinate system (ECEF). The position of the signal when it propagates to the receiver can be extrapolated by the GNSS satellite orbit. In terms of constraints, the earth's surface is divided into land and ocean, so the mean sea surface (MSS) model is used for ocean reflection, such as DTU13. Land reflection uses terrain data, such as the Shuttle Radar Topography Mission (SRTM).

[0061] Step 13: Construct a Doppler model based on the mirror reflection point position and the reflection timestamp, and generate a local carrier I / Q signal (local replica signal) based on the Doppler model; the Doppler model is a model of the reflection geometry constructed by the mirror reflection point, including two parameters: the time delay and Doppler frequency of the reflected signal.

[0062] Step 14: Obtain the I / Q data segments of the H and V channels according to the GNSS-R data set, and perform correlation processing (convolution) on the I / Q data segments and the local carrier I / Q signals to generate a delay Doppler map (DDM).

[0063] The present invention provides a polarization GNSS-R receiver algorithm and a polarization calibration algorithm for a delay Doppler map DDM through the above scheme; wherein the polarization GNSS-R receiver algorithm includes: position solution of direct signals, prediction of mirror reflection points, construction of Doppler delay model and design of reflection channel DDM processor; the receiver antenna receives GNSS reflection signals in a linear polarization mode, the data is processed in H and V channels, and a complex delay Doppler map DDM of the corresponding channel is generated, which is convenient for the subsequent generation of full Stokes parameters; the polarization calibration algorithm of the delay Doppler map DDM is In the quasi-algorithm, based on the delay Doppler map DDM generated by the polarization GNSS-R receiver algorithm, the Stokes parameters are used to characterize the polarization characteristics of the reflected signal. On the basis of the Stokes parameters, the present invention proposes three non-ideal effects that need to be calibrated (polarization calibration of the receiver, calibration of two Faraday rotations (FAR) on the reflection path, and calibration of the transmitter signal axis ratio), and then converts the calibration factors of different non-ideal effects into compensation matrices of the same dimension by constructing a matrix. Finally, the Stokes parameters are compensated by multiplication to complete the calibration of the DDM data. Compared with the prior art, the present invention has the following beneficial effects:

[0064] 1. Based on the data acquired by the polarization GNSS-R receiver, the present invention uses Stokes parameters to calibrate the existing polarization non-ideal effects, thereby solving the gain loss problem caused by the difference between the polarization direction of the receiver antenna and the polarization direction of the reflected signal, the phase shift problem caused by the bidirectional Faraday rotation of the reflected signal after passing through the ionosphere twice, and the non-ideal state problems such as elliptical and phase shift caused by the non-ideal axial ratio of the GNSS transmitting antenna;

[0065] 2. The algorithm proposed in the present invention can obtain the axis ratio information of the corresponding GNSS satellite, which is convenient for application in the subsequent product generation process to improve the accuracy of the product;

[0066] 3. At the same time, the algorithm of the present invention calibrates the polarization direction of the antenna to a certain extent, increases the power of the reflected signal, and reduces the antenna load on the payload. BRIEF DESCRIPTION OF THE DRAWINGS

[0067] The present invention may be better understood by referring to the following description given in conjunction with the accompanying drawings, wherein the same or similar reference numerals are used throughout the drawings to represent the same or similar components. The accompanying drawings, together with the following detailed description, are included in and form a part of this specification and are used to further illustrate the preferred embodiments of the present invention and to explain the principles and advantages of the present invention. In the drawings:

[0068] Figure 1 This is a flow chart of the polarization GNSS-R receiver algorithm of the present invention;

[0069] Figure 2 It is the reflection channel process of circular polarization antenna in conventional GNSS-R receiver;

[0070] Figure 3 The reflection channel process of the GNSS-R linear polarization receiver of the present invention;

[0071] Figure 4 This is a flow chart of signal polarization calibration of the present invention;

[0072] Figure 5 A general processing scheme for obtaining Stokes parameters for complex DDM;

[0073] Figure 6 A flow chart of obtaining two DDM diagrams in the polarization GNSS-R receiver algorithm of the present invention;

[0074] Figure 7 Schematic diagram of receiver polarization calibration. DETAILED DESCRIPTION

[0075] Embodiments of the present invention will be described below with reference to the accompanying drawings. Elements and features described in one drawing or one embodiment of the present invention may be combined with elements and features shown in one or more other drawings or embodiments. It should be noted that for the purpose of clarity, the representation and description of components and processes that are not related to the present invention and are known to those of ordinary skill in the art are omitted in the drawings and descriptions.

[0076] The purpose of the present invention is to design a calibration algorithm after the polarization GNSS-R receiver data is processed to calibrate the non-ideal factors from the signal and the payload, including the receiver effect, the scene antenna polarization misalignment, the Faraday rotation and the transmitter non-ideality, etc., and to compensate the product one by one by obtaining the corresponding parameter values, so as to obtain a higher quality GNSS-R product. In the prior art, the above non-ideal factors have affected the polarization radar receiver and the GNSS radio occultation receiver and have been solved, but the GNSS-R receiver has not.

[0077] First, the present invention provides an algorithm flow of a polarization GNSS-R receiver. Figure 1 As shown, the steps of the polarization GNSS-R receiver algorithm are as follows:

[0078] The first step is to generate GNSS-R data sets, including I / Q data sets received by the direct antenna, GNSS satellite PVT (position, velocity, time) information and satellite clock error, and I / Q data sets of H and V channels received by the reflection antenna. The receiver uses two orthogonal linear polarization receiving antennas, and can use a pair of circular polarization antennas with opposite meanings to receive vertical polarization, horizontal polarization, and cross-polarization reflected signals.

[0079] The second step is to find the mirror reflection point position and reflection timestamp. The mirror reflection point position (SP) is estimated by finding the following function that minimizes the reflection path length, while satisfying the condition that the incident angle at the mirror reflection point is equal to the reflection angle, and then the mirror reflection point position is constrained to the earth's surface. The following function is shown as follows:

[0080]

[0081] In the formula are the position vectors of the GNSS-R receiver, the mirror reflection point, and the GNSS satellite in the Earth-centered Earth-fixed rectangular coordinate system (ECEF). The position of the signal when it propagates to the receiver can be extrapolated by the GNSS satellite orbit. In terms of constraints, the earth's surface is divided into land and ocean, so the mean sea surface (MSS) model is used for ocean reflection, such as DTU13. Land reflection uses terrain data, such as the Shuttle Radar Topography Mission (SRTM).

[0082] The third step is to obtain the I / Q data segments and Doppler models of the H and V channels respectively. The data segments are used for open-loop tracking of the reflected signal and generation of the DDM diagram. The Doppler model is a model of the reflection geometry constructed by the mirror reflection points, including two parameters: the delay and Doppler frequency of the reflected signal. This model can assist the receiver in building a local replica signal.

[0083] Step 4: Obtain a Delay Doppler Map (DDM). Correlate the local replica signal with the H channel data fragment (horizontally polarized data) and V channel data fragment (vertically polarized data) obtained in step 3 and save the results separately, in the same way as a conventional reflection receiver.

[0084] Among them, the construction of the Doppler delay model in the third step includes the following process:

[0085] Using the master-slave scheme, the code delay and Doppler frequency of the reflected signal are denoted as τ R (t) and f R (t), according to the direct signal code delay τ T (t), Doppler frequency f T(t), the mirror reflection point SP is used to model it:

[0086]

[0087]

[0088] Where R T (t) and Represents the distance and rate of change of the direct signal path, R R (t) and represents the distance and distance change rate of the reflected signal path, δt is the time delay between the direct and reflected signals, Cps is the chip rate, and f c is the carrier frequency of the GNSS signal, c is the propagation speed of electromagnetic waves;

[0089] Direct signal code delay τ T (t), Doppler frequency f T (t), distance change rate It can be obtained by capturing and tracking the direct signal; the reflected path R R (t) and It can be calculated based on the position of the specular reflection point SP. The determination of the specular reflection point needs to be combined with geographic information. At present, most specular reflection points are determined based on the general empirical model WGS84 coordinate system. In order to obtain more detailed specular reflection points, the following operations can also be performed: distinguish the environment of the reflection point. For example, if the reflection point is an ocean surface environment, the mean sea surface model (MSS) is used, such as DTU13. If the reflection point is a land environment, terrain data is used, such as the Shuttle Radar Topography Mission (SRTM). Both of these measured terrain models are existing technologies and will not be described in detail here.

[0090] In order to further improve the acquired delay Doppler map DDM, the GNSS-R receiver of the present invention is different from the existing receiver in terms of antenna and data reflection channel. The conventional GNSS-R receiver is a circularly polarized antenna, which receives the I-path and Q-path composite signals (I+Qj) from one signal. In this regard, the conventional GNSS-R reflection channel process (one path) is generally 1ms, generating 1 DDM, and its principle diagram can be seen in Figure 2 shown.

[0091] See also Figure 3 The reflection channel process (two paths) of the GNSS-R linear polarization receiver of the present invention generates two DDMs (H channel data segments DDM) in the same time. H and V channel data fragment DDM V). Therefore, the present invention extracts two characteristics, horizontal polarization and vertical polarization, to study the polarization characteristics of the signal, so as to facilitate the subsequent algorithm to correct it, such as elliptical calibration and polarization direction calibration. In order to adapt to the two-way reflection channel, the GNSS-R receiver of the present invention adopts a linear polarization polarization receiver, and the antenna is divided into horizontal polarization and vertical polarization, which respectively receive signals from the horizontal and vertical directions, that is, two-way signals; there are two processing modules, which process the two-way signals respectively. When processing the reflection channel of a conventional receiver, it is a data carrier demodulation time-division I / Q channel. When receiving, the polarization receiver of the present invention will generate two data, which need separate carrier demodulation and then generate two DDMs at the end. These two DDMs represent signal information in different polarization directions, and may also contain information from the ground. Because conventional antennas are circularly polarized, no one in the prior art has designed two processing modules for the reflection channel; in order to process dual linear polarization, the present invention innovatively adopts two processing modules to distribute and extract polarization characteristics for separate processing.

[0092] In addition, in the above-mentioned GNSS-R receiver algorithm, the direct signal position solution and mirror reflection point prediction use the general GNSS navigation receiver internal position solution algorithm and mirror reflection point prediction algorithm and DDM processor, which are not the invention points of the present invention and will not be described in detail here.

[0093] The calibration algorithm provided by the present invention can be applied to the DDM data processed by the polarization GNSS-R receiver. In the algorithm scheme, a polarization scenario of GNSS reflection is first defined to describe the polarization characteristics of the reflected signal. Then, different non-ideal effects that need to be compensated are defined, and the corresponding calibration matrix is ​​obtained by analyzing various non-ideal effects to compensate for the polarization scenario, and finally the non-ideal calibration of the DDM diagram is achieved. Specifically, the polarization calibration method of the polarization GNSS-R receiver of the embodiment of the present invention has an algorithm flow chart as shown below: Figure 4 The detailed steps are as follows:

[0094] Step 1: Build a polarization scenario. Before calibrating non-idealities, it is necessary to extract the polarization characteristics of the GNSS reflection signal, namely the Stokes parameters. The Stokes parameters are parameters that describe the polarization state of electromagnetic waves. They can replace the total intensity (I) and polarization (p) of polarized radiation, as well as the shape parameters of the polarization ellipse. The Stokes parameters are used to further invert the specific parameters of the decomposition, such as the degree of polarization (DoP), relative phase (d) and ellipticity parameters (x). The Stokes parameters can be obtained from the complex DDM of the H and V polarization channels, which is called the full Stokes DDM, which is the product generated by the polarization GNSS-R receiver mentioned above. Figure 5 A general solution for obtaining Stokes parameters from complex DDM is given. The formula for extracting Stokes parameters is:

[0095] S0=<|E RH | 2 >+<|E RV | 2 >

[0096] S1=<|E RH | 2 >-<|E RV | 2 >

[0097]

[0098]

[0099] Where: E RH and E RV It is the complex electric field after the two linear orthogonal polarization components (H and V) of the right-hand circularly polarized transmitted signal are cross-correlated with the C / A code replica.

[0100] See also Figure 6 , two DDM diagrams (DDM H and DDM V ), these two types of DDM are obtained through different polarization channels, and the DDM diagram is a two-dimensional matrix. Figure 6 It is assumed that no incoherent integration is performed and only 1ms signal is extracted for coherent integration. That is to say, the process of the polarization GNSS-R receiver algorithm is only executed once, and two DDM diagrams are generated at a time. Figure 1 There are N values ​​in total, and four Stokes parameters can be extracted from the two DDM diagrams. This process is the construction of the polarization scenario in step 1; each parameter obtained is a sub-parameter of the Stokes parameter. The present invention applies the Stokes parameter to CYGNSS data or polarization receiver data for the first time.

[0101] Step 2: define different non-ideal effects that need to be compensated. The present invention provides three non-ideal effects that need to be calibrated, namely, polarization calibration of the receiver, calibration of two Faraday rotations (FAR) on the reflection path, and calibration of the axial ratio of the transmitter signal.

[0102] Step 3: calibrate the non-ideal effects proposed in step 2. All calibration effects will act on the Stokes parameters, and the extracted Stokes parameters will be calibrated. After calibration, the calibrated Stokes parameters are used to calculate the set indicators, such as reflectivity, etc., which is the calibration of the DDM diagram.

[0103] The calibration methods provided by the present invention are all based on the Stokes parameters defined in step one. In step four, the factors that need to be compensated for different non-ideal effects are converted into error matrices of the same dimension. Therefore, only the respective calibration methods are described in this step.

[0104] (1) Receiver polarization calibration: It is necessary to calculate the pointing direction of the axis of the receiver antenna.

[0105] δθ=θ sp -θ T

[0106]

[0107] where θ T and are the azimuth and elevation angles of the actual pointing direction of the receiver antenna, θ sp and are the azimuth and elevation angles at the mirror reflection point respectively. The angle calculation uses the Earth-centered Earth-fixed coordinate system.

[0108] δθ and It is the azimuth and elevation angle offset in the clockwise direction with the receiver position as the origin and the mirror reflection point as the target. The angle calculation uses the Earth-centered Earth-fixed coordinate system.

[0109] In the prior art, the non-ideal effect of the receiver is actually because the receiving antenna is not aligned with the direction of the mirror reflection point when receiving the signal. One disadvantage of the linear polarization antenna used in the present invention is that it must be aligned with the receiving point when receiving the signal, so that the signal can be received completely. If it is not aligned, it will lead to low power of the received signal and a certain phase offset. Therefore, for the polarization calibration of the receiver, it is sufficient to calculate the angle of the antenna offset. Figure 7 , the offset calculation involves two angles, one is the azimuth angle and the other is the elevation angle. These two angles are calculated and then calibrated in combination with the receiver non-ideal effect calibration method.

[0110] (2) Faraday rotation calibration: Faraday rotation will cause the polarization direction of the reflected signal to be different from the polarization direction of the receiver antenna. Therefore, the Faraday rotation angle needs to be estimated during calibration. The estimation method is as follows:

[0111] Ω≈λ 2 TEC U

[0112] Where λ is the wavelength of the reflected signal, TEC U is the normalized total column electron content, obtained by inverting the TEC map provided by GPS.

[0113] (3) Calibration of the transmitter signal axial ratio: To calculate the full Stokes parameters of the linearly polarized antenna, it is necessary to consider the axial ratio of the GNSS satellite transmission. The calibration of the axial ratio needs to be obtained through machine learning fitting. Specifically, 1. Use the GNSS-R L1 and ERA5 data sets to invert the full polarization model of the reflected signal. 2. Perform a preliminary calibration of the Stokes parameters obtained in step 1, including Faraday rotation calibration and receiver polarization calibration. 3. Set the optimizer and loss function. Here is a minimized loss function, as shown below:

[0114]

[0115] Where S n (m)(n=0,1,2,3) is the full polarization model with unknown parameter m, is the corrected Stokes parameter.

[0116] Step 4: Convert the factors that need to be compensated for the different non-ideal effects in step 3 into error matrices of the same dimension. Here, we first define a matrix A to describe the error factors of the receiving antenna and the transmitting antenna.

[0117]

[0118] Where m is the antenna axial ratio, is the phase difference between H and V polarizations. σ1 and σ2 are the normalized powers of H and V polarizations, respectively.

[0119] Define the matrix C as the error factor describing the Faraday rotation.

[0120]

[0121] Where Ω is the Faraday rotation angle obtained in step 3

[0122] Step five, calibration of Stokes parameters, the calibration is shown in the following formula.

[0123]

[0124] In the formula and S n are the calibrated and uncalibrated Stokes parameters, n=0, 1, 2, 3 respectively. and are the calibration matrices for the transmitter and receiver, respectively, C -1 is the calibration matrix for the Faraday rotation.

[0125] Specifically, the calibration process is to first construct a matrix formula for the Stokes parameters, and then multiply the Stokes parameters by the inverse matrix of the error factor.

[0126] More specifically, first construct a matrix formula:

[0127] S n =A T ·C·A R ·S FSA

[0128] Among them A T , C, A R are the error factors of the transmitting antenna, Faraday rotation, and receiving antenna, respectively, S FSA For the forward scattering alignment base (existing technology), then S n The (S0, S1, S2, S3) parameters are composed of error factors and basis.

[0129] Then multiply the Stokes parameters by the inverse matrix of the error factor (S n =A T ·C·A R ·S FSA ) to get the calibrated Stokes parameters The inverse matrices of the error factors are C -1 ,

[0130] After obtaining the modified Stokes parameters, the corresponding indicators can be calculated, such as calculating the surface reflectivity <|E RV | 2 >, surface reflectivity is used as the main observation formula for soil inversion;

[0131]

[0132] Step 6: Invert and restore the calibrated Stokes parameters into a DDM map to achieve calibration of the DDM map.

[0133] The present invention adopts the above scheme, uses the processed DDM to replace the I / Q polarization data, and uses the satellite processed DDM as the calibration input by comparing the I / Q data stream and generating the DDM, which has a lower data rate and a higher signal-to-noise ratio, and is convenient for polarization calibration. Although the calibration scheme is designed based on a linear polarization GNSS receiver, it is also applicable to a general circular polarization GNSS-R receiver. The other undescribed parts of the GNSS-R receiver algorithm of the present invention are roughly the same as the algorithms in conventional receivers, and the conventional receiver algorithms can be referred to.

[0134] The method of the present invention is not limited to being executed in the time sequence described in the specification, and may also be executed in other time sequences, in parallel or independently. Therefore, the execution order of the method described in this specification does not limit the technical scope of the present invention.

[0135] Although the present invention has been disclosed above by describing specific embodiments of the present invention, it should be understood that all the above embodiments and examples are exemplary rather than restrictive. Those skilled in the art may design various modifications, improvements or equivalents of the present invention within the spirit and scope of the appended claims. These modifications, improvements or equivalents should also be considered to be included in the protection scope of the present invention.

Claims

1. A polarization calibration method for a polarization GNSS-R receiver, characterized in that: include: Step 1: Obtain the delay Doppler map DDM; Obtain Stokes parameters from the delay Doppler map DDM; Step 2: Define one or more non-ideal effects that need to be compensated. Step 3: Calibrate the non-ideal effects defined in step 2 based on the Stokes parameters; the non-ideal effects include polarization calibration of the receiver, two Faraday rotation calibrations on the reflection path, and calibration of the transmitter signal axial ratio; Step 4: Convert the factors that need to be compensated for the different non-ideal effects in step 3 into error matrices of the same dimension; Step 5: Calibrate the Stokes parameters based on the error matrix of step 4; Step 6: Invert and restore the calibrated Stokes parameters into a delay-Doppler map (DDM map).

2. The polarization calibration method for a polarization GNSS-R receiver according to claim 1, characterized in that: In the step 1, the Stokes parameters are obtained from the delay Doppler map DDM, specifically including: constructing a polarization scenario, extracting the polarization characteristics of the GNSS reflection signal, and obtaining the Stokes parameters; the Stokes parameters are parameters that describe the polarization state of electromagnetic waves, and the formula for extracting the Stokes parameters is: S0=<|E RH | 2 >+<|E RV | 2 >; S1=<|E RH | 2 >-<|E RV | 2 >; Where: S0-S3 are Stokes parameters, E RH and E RV It is the complex electric field after the two linear orthogonal polarization components H and V of the right-hand circularly polarized transmitted signal are cross-correlated with the C / A code replica.

3. The polarization calibration method for a polarization GNSS-R receiver according to claim 1, characterized in that: The polarization calibration of the receiver based on the Stokes parameters in step 3 specifically includes: Calculate the direction in which the axis of the receiver antenna is pointing: δθ=θ sp -θ T ; where θ T and are the azimuth and elevation angles of the actual pointing direction of the receiver antenna, θ sp and are the azimuth and elevation angles at the mirror reflection point, respectively. The angle calculation uses the Earth-centered Earth-fixed coordinate system; δθ and It is the azimuth and elevation angle offset in the clockwise direction with the receiver position as the origin and the mirror reflection point as the target. The angle calculation uses the Earth-centered Earth-fixed coordinate system.

4. The polarization calibration method for a polarization GNSS-R receiver according to claim 3, characterized in that: In step 3, calibrating the two Faraday rotations on the reflection path based on the Stokes parameters specifically includes: Estimate the Faraday rotation angle Ω, the estimation method is as follows: Ω≈λ 2 ·TEC U ; Where λ is the wavelength of the reflected signal, TEC U is the normalized total column electron content, obtained by inverting the TEC map provided by GPS.

5. The polarization calibration method for a polarization GNSS-R receiver according to claim 4, characterized in that: In step 3, calibrating the axial ratio of the transmitter signal based on the Stokes parameters specifically includes: using GNSS-R L1 level data and ERA5 data set to invert the full polarization model of the reflected signal; Performing a preliminary calibration on the Stokes parameters obtained in step 1; Set up the optimizer and loss function to obtain the calibrated Stokes parameters.

6. The polarization calibration method for a polarization GNSS-R receiver according to claim 5, characterized in that: The loss function to be minimized is as follows: Where m is the antenna axial ratio, S n (m) is the full polarization model with unknown parameter m, is the corrected Stokes parameter, n=0,1,2,3.

7. The polarization calibration method for a polarization GNSS-R receiver according to claim 4, characterized in that: Convert the factors that need to be compensated for different non-ideal effects in step 3 into error matrices of the same dimension, including: Define a matrix A to describe the error factors of the receiving antenna and the transmitting antenna: Where m is the antenna axial ratio, is the phase difference between H and V polarizations; σ1 and σ2 are the normalized powers of H and V polarizations, respectively; e jφ =cos(φ)+jsin(φ); Define the matrix C as the error factor describing the Faraday rotation: Where Ω is the Faraday rotation angle obtained in step 3.

8. The polarization calibration method for a polarization GNSS-R receiver according to claim 7, characterized in that: In step 5, the calibration of Stokes parameters is shown as follows: In the formula and S n are the calibrated and uncalibrated Stokes parameters, n=0, 1, 2, 3; and are the calibration matrices for the transmitter and receiver, respectively, C -1 is the calibration matrix for the Faraday rotation.

9. The polarization calibration method for a polarization GNSS-R receiver according to claim 1, characterized in that: The step 1: obtaining the delay Doppler map DDM specifically includes: Step 11: Generate a GNSS-R dataset, which includes a dataset received by a direct antenna and a dataset received by a reflection antenna. The dataset received by the direct antenna includes: an I / Q dataset, GNSS satellite PVT information, and satellite clock error received by the direct antenna; the dataset received by the reflection antenna includes an I / Q dataset of H and V channels received by the reflection antenna; Step 12: Calculate the mirror reflection point position and reflection timestamp based on the GNSS-R data set. Specifically, the mirror reflection point position is calculated by finding a following function that minimizes the reflection path length and satisfies the condition that the incident angle at the mirror reflection point position is equal to the reflection angle. Then, the mirror reflection point position is constrained to the earth's surface. The following function is as follows: In the formula They are the position vectors of the GNSS-R receiver, the mirror reflection point, and the GNSS satellite in the Earth-centered Earth-fixed rectangular coordinate system; Step 13: Construct a Doppler model based on the mirror reflection point position and the reflection timestamp, and generate a local carrier I / Q signal based on the Doppler model; the Doppler model is a model of reflection geometry constructed by the mirror reflection point, including two parameters: the time delay and Doppler frequency of the reflected signal; Step 14: Obtain the I / Q data segments of the H and V channels according to the GNSS-R data set, and perform correlation processing on the I / Q data segments and the local carrier I / Q signals to generate a delay Doppler map DDM.

Images