A space-based calibration ground clutter suppression method for active microwave remote sensing satellite

Abstract

The application belongs to the technical field of satellite calibration, and discloses a space-based calibration ground clutter suppression method for an active microwave remote sensing satellite, which comprises the following steps: step S1, a Doppler term in a slow time domain is constructed by using the difference between the ground speed of the active microwave remote sensing satellite relative to a space-based calibration satellite; step S2, different manifold space matrices are formed based on the Doppler term in the slow time domain; and step S3, a subspace is constructed by using the manifold matrix, subspace projection is performed, and the expected signal is recovered. The space-based calibration ground clutter suppression method for the active microwave remote sensing satellite has the advantages of no need for additional design and hardware cost, low system complexity and operation cost, and fast error convergence.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of satellite calibration, and particularly relates to a space-based calibration ground clutter suppression method for an active microwave remote sensing satellite. BACKGROUND

[0002] System calibration is the basis for reliable operation of space-based microwave remote sensing, and its application needs to meet the two core requirements of "real-time calibration" and "global coverage". Influenced by space radiation environment and equipment aging, the performance of microwave sensors presents dynamic change characteristics, and "anytime and anywhere" calibration has become the key to maintaining high-performance operation of the system. Traditional ground calibration methods cannot meet the calibration needs of polar regions, remote seas and other areas. Therefore, developing calibration technology based on space platforms and building a global calibration network on satellites is the premise for achieving all-weather and all-region high-precision observation tasks.

[0003] The space-based calibration satellite can avoid atmospheric interference by calibrating the target star outside the atmosphere, and can also achieve high-frequency calibration by designing the orbit. However, the active microwave remote sensing satellite being calibrated will simultaneously receive signals from the calibration satellite and clutter from the ground, which has a non-negligible impact on the accuracy of calibration.

[0004] Current interference suppression techniques mainly rely on space-time signal processing and coherent pulse accumulation. The space-time signal processing method mainly uses the spatial distribution characteristics and time evolution characteristics of the signal. After sufficient sampling, an inversion clutter covariance matrix is constructed. This method requires at least twice the number of independent and identically distributed samples of the system; at the same time, the clutter covariance matrix obtained by space-time signal processing is limited to the sampling space-time location, and the timeliness is limited; and the additional sampling also increases the boot-up time and energy consumption. The coherent pulse accumulation method is still limited to specific land / sea areas, and the method assumes that the clutter phase changes slowly in the azimuth angle, which may not effectively suppress the clutter for high-speed moving satellites.

[0005] Therefore, in view of the above technical defects, the present application provides a space-based calibration ground clutter suppression method for an active microwave remote sensing satellite. SUMMARY

[0006] The purpose of the present application is to provide a space-based calibration ground clutter suppression method for an active microwave remote sensing satellite, which utilizes the different speeds of the active microwave remote sensing satellite relative to the ground and the calibration satellite, forms different orthogonal projection spaces after orbiting, and then suppresses the ground clutter in the received signal through the orthogonal projection spaces.

[0007] To achieve the above purpose, the present application provides a space-based calibration ground clutter suppression method for an active microwave remote sensing satellite, comprising the following steps:

[0008] Step S1, using the difference between the ground speed of the active microwave remote sensing satellite and the space-based calibration star, a Doppler term in the slow time domain is constructed;

[0009] Step S2, based on the Doppler term in the slow time domain, different manifold space matrices are formed;

[0010] Step S3, using the manifold matrix to construct a subspace, subspace projection is performed to recover the desired signal.

[0011] Preferably, in step S1, the radar transmitted linear frequency modulation (LFM) signal The expression is:

[0012] (1);

[0013] Wherein, is an exponential function; is the fast time; is the pulse duration; is the imaginary unit; is the carrier frequency; is the frequency modulation.

[0014] Preferably, the echo signal from the ground clutter The expression at time is:

[0015] (2);

[0016] Wherein, is the slow time; is the amplitude of the ground clutter; is the distance of the active microwave remote sensing satellite relative to the ground; is the speed of light; is the wavelength; is the speed of the active microwave remote sensing satellite relative to the ground.

[0017] Preferably, the echo signal from the calibration star The expression at time is:

[0018] (3);

[0019] Wherein, is the echo amplitude of the calibration star; is the distance of the active microwave remote sensing satellite relative to the calibration star; is the speed of the active microwave remote sensing satellite relative to the calibration star.

[0020] The overall received signal of the active microwave remote sensing satellite is preferably written as a form containing ground clutter, signal of space-based calibration star and noise, which is at the expression of time is:

[0021] (4);

[0022] wherein, is a Gaussian white noise;

[0023] The overall received signal of the active microwave remote sensing satellite is written as a matrix form of dimension, denoted as ; wherein, is the sampling number of slow time, is the sampling number of fast time.

[0024] Preferably, in step S2, according to the orbit information, the distance and the speed of the active microwave remote sensing satellite relative to the ground are obtained, the distance and the speed of the active microwave remote sensing satellite relative to the calibration star are obtained;

[0025] According to the speed information, a vector containing Doppler terms is formed, , as shown below:

[0026] (5);

[0027] Let , except for the Doppler term containing slow time , the remaining part is denoted as a matrix of dimension , as shown below:

[0028] (6);

[0029] (7);

[0030] wherein, is an item from the ground containing only fast time echo information; is an item from the calibration star containing only fast time echo information; is the sampling of each time of the fast time echo from the ground; is the sampling of each time of the fast time echo from the calibration star;

[0031] is obtained by matrix multiplication; wherein, ​is a noise matrix.

[0032] Preferably, according to , forming a manifold space matrix with As shown below:

[0033] (8);

[0034] wherein, is a manifold space matrix containing ground echo information spanned by column vectors of ; is a manifold space matrix containing calibration star echo information spanned by column vectors of .

[0035] Preferably, in step S3, a subspace is constructed by using the manifold matrix, a subspace projection is performed, and the desired signal is recovered, and the specific process is as follows:

[0036] The received signal is projected to the manifold space matrix, and since the speed of the active microwave remote sensing satellite relative to the ground is inconsistent with the speed of the active microwave remote sensing satellite relative to the calibration star, the complex exponential term on and is irrelevant, and the inner product is zero, so:

[0037] (9);

[0038] According to the above formula, it is further obtained that:

[0039] (10);

[0040] Since the noise is irrelevant to the signal, that is, so the estimated echo signal from the calibration star is obtained as shown below:

[0041] (11).

[0042] Therefore, the space-based calibration ground clutter suppression method for the active microwave remote sensing satellite has the beneficial effects as follows:

[0043] 1) The present application only relies on satellite orbital speed information without the need for additional hardware support; since most satellites already have precise orbit determination modules and speed measurement capabilities, direct application is not necessary without system modification or additional expense, and without relying on additional sampling time for clutter statistical estimation, there is no need for special waveform design, and it has good universality and practicality.

[0044] 2) The application realizes decoupling of the processing flow and the satellite platform; all clutter suppression processing can be completed on the ground through the data transmission link, avoiding dependence on on-board processing capacity; at the same time, the scheme of the application does not require inter-satellite communication and cooperative operation, significantly simplifying the calibration and operation flow, reducing system complexity and operation cost.

[0045] 3) The application realizes fast error convergence by constructing a high-precision projection subspace through multiple pulses; with the increase of the number of pulses, the precision of the projection space estimation is steadily improved, thereby effectively suppressing clutter interference and improving overall system performance.

[0046] The technical solutions of the application will be further described in detail below with reference to the drawings and embodiments. BRIEF DESCRIPTION OF DRAWINGS

[0047] Figure 1 is a flowchart of a space-based calibration ground clutter suppression method for an active microwave remote sensing satellite according to the application;

[0048] Figure 2 is a change relationship of MSE with respect to the dimension of the subspace according to the application;

[0049] Figure 3 is a change relationship of MSE with respect to SNR according to the application;

[0050] Figure 4 is a change relationship of MSE with respect to SCR according to the application. DETAILED DESCRIPTION

[0051] The technical solutions of the application will be further described in detail below with reference to the drawings and embodiments.

[0052] As shown in the drawings, Figure 1 a space-based calibration ground clutter suppression method for an active microwave remote sensing satellite according to the application includes the following steps:

[0053] Step S1, a Doppler term in the slow time domain is constructed by utilizing the difference in speed of the active microwave remote sensing satellite with respect to the space-based calibration satellite and the ground.

[0054] Step S2, different manifold space matrices are formed based on the Doppler term in the slow time domain.

[0055] Step S3, a subspace is constructed using the manifold matrix, subspace projection is performed, and the desired signal is recovered.

[0056] Embodiment 1

[0057] Step S1, a Doppler term in the slow time domain is constructed by utilizing the difference in speed of the active microwave remote sensing satellite with respect to the space-based calibration satellite and the ground.

[0058] ​Step S11, the expression of the Linear Frequency Modulated (LFM) signal transmitted by the radar is:

[0059] (1);

[0060] wherein, is an exponential function; is a slow time; is a pulse duration; is an imaginary unit; is a carrier frequency; is a frequency modulation rate.

[0061] Step S12, the expression of the echo signal from the ground clutter at time

[0062] (2);

[0063] wherein, is a slow time; is an amplitude of the ground clutter; is a distance of the active microwave remote sensing satellite relative to the ground; is a speed of light; is a wavelength; is a speed of the active microwave remote sensing satellite relative to the ground.

[0064] Step S13, the expression of the echo signal from the calibration satellite at time

[0065] (3);

[0066] wherein, is an echo amplitude of the calibration satellite; is a distance of the active microwave remote sensing satellite relative to the calibration satellite; is a speed of the active microwave remote sensing satellite relative to the calibration satellite.

[0067] Step S14, the overall received signal of the active microwave remote sensing satellite is written as a form containing the ground clutter, the signal of the space-based calibration satellite and noise, and the expression of the overall received signal of the active microwave remote sensing satellite at time

[0068] (4);

[0069] wherein, is a Gaussian white noise.

[0070] ​​​​​​​For convenience of calculation, it is written as In matrix form, denoted as ; wherein, is the sampling number of slow time, is the sampling number of fast time.

[0071] Step S2, based on the Doppler term on the slow time domain, form different manifold space matrices.

[0072] Step S21, according to the orbit information, get the distance and velocity of the active microwave remote sensing satellite relative to the ground. and velocity of the active microwave remote sensing satellite relative to the calibration star.

[0073] Step S22, according to the velocity information, form a vector containing , Doppler term, as follows:

[0074] (5);

[0075] Let , except for the Doppler term containing the slow time , the rest is denoted as a matrix with dimension , as follows:

[0076] (6);

[0077] (7);

[0078] wherein, is the term from the ground containing only fast time echo information; is the term from the calibration star containing only fast time echo information; is the sampling of each time from the ground fast time echo; is the sampling of each time from the calibration star fast time echo.

[0079] Through matrix multiplication, get ; wherein, is the noise matrix.

[0080] Step S23, according to , form manifold space matrix and , as follows:

[0081] ​(8);

[0082] in, For the reason The column vectors spanned by the manifold space matrix containing ground echo information; For the reason The column vectors span a manifold space matrix containing calibration star echo information.

[0083] Step S3: Construct a subspace using a manifold matrix, project the subspace, and recover the desired signal.

[0084] Projecting the received signal onto the manifold space matrix, since the velocity of the active microwave remote sensing satellite relative to the ground is always different from the velocity of the active microwave remote sensing satellite relative to the calibration satellite, therefore... and The complex exponential terms on the x-axis are uncorrelated, and their inner product is also zero, therefore we can obtain:

[0085] (9);

[0086] Based on the above formula, we can further obtain:

[0087] (10);

[0088] Since noise is uncorrelated with the signal, therefore... This allows us to obtain an estimated echo signal from the calibration star. As shown below:

[0089] (11);

[0090] Based on the above process, ground clutter can be effectively suppressed from the echo by utilizing velocity differences, and the signal of the calibration satellite can be extracted.

[0091] Example 2

[0092] This embodiment further illustrates the beneficial effects of the present invention through experimental results. The software used in the experiment is MATLAB, and the echo data of the spaceborne C-band system is used to verify the ground clutter suppression performance.

[0093] The experimental conditions were: carrier frequency 5.4 GHz, bandwidth 50 MHz, sampling rate 100 MHz, pulse repetition frequency 10 kHz, active microwave remote sensing satellite relative to the ground velocity 7 km / s, and relative ground velocity 4.2 km / s.

[0094] In subsequent simulations, the amplitude of the received signal from the calibration satellite was fixed at 1. To analyze the effectiveness of the proposed algorithm in clutter suppression, [further details are needed]. Simulation experiments were conducted to measure the dimension, signal-to-noise ratio (SNR), and signal-to-noise ratio (SCR).

[0095] The mean square error (MSE) is introduced as a performance indicator as follows:

[0096] (12);

[0097] wherein, denotes the number of simulations, denotes the number of sampling points within each pulse.

[0098] (1) The clutter suppression performance varies with the dimension of .

[0099] In the scheme proposed in the application, the inner product operation of is introduced, thereby ensuring the orthogonality between the ground clutter and the projection subspace. Meanwhile, the higher the dimension of the vector, the stronger the stability of the orthogonality. As shown in Figure 2 , the dimension of varies from 5 to 35, the SNR is set to 10 dB, and the SCR is 0 dB. The results show that, as the dimension of increases, the MSE significantly decreases; when the dimension is 10, the MSE is close to -20 dB, indicating that the scheme has good estimation accuracy.

[0100] (2) The clutter suppression performance varies with the SNR.

[0101] Under the premise that the signal and the noise are uncorrelated and the noise is independent and identically distributed (IID), the noise model is , that is, a complex Gaussian distribution with zero mean and covariance . As shown in Figure 3 , the SNR varies from 0 to 10, the dimension is 10, and the SCR is 0 dB. It can be observed that the MSE decreases approximately linearly as the SNR increases. This is because the white noise is uniformly distributed in all dimensions, and the application focuses on the dimension of the received signal. The error in this dimension is mainly affected by the noise in this dimension, and the final MSE is approximately equal to the variance of the noise in this dimension.

[0102] (3) The clutter suppression performance varies with the SCR.

[0103] As shown in Figure 4 , the SCR varies from 0 to 10, the dimension is 10, and the SNR is 10 dB. It can be seen that the MSE in this simulation is lower than the result under the SNR variation condition. This is because, unlike the noise, the clutter signal is concentrated in a limited dimension, and the application successfully suppresses the clutter components in these dimensions through the orthogonality of the manifold matrix, further indicating that the method has good effect in clutter suppression.

[0104] Therefore, the application adopts the above-mentioned ground clutter suppression method for space-based calibration of an active microwave remote sensing satellite, introduces the relative speed difference between the active microwave remote sensing satellite and the calibration satellite and the ground, and uses it to construct a subspace projection matrix; by projecting the received signal into the subspace, the ground clutter signal can be effectively suppressed, and the required target signal can be extracted. Compared with the traditional method, the scheme has the advantages of no additional design and hardware overhead, lower system complexity and operation cost, and fast error convergence.

[0105] Finally, it should be noted that: the above examples are only used to illustrate the technical solutions of the present application, but not to limit them, although the present application has been described in detail with reference to the preferred embodiments, those skilled in the art should understand that: it can still modify or equivalently replace the technical solutions of the present application, and these modifications or equivalent replacements also cannot make the modified technical solutions deviate from the spirit and scope of the technical solutions of the present application.

Claims

1. A space-based calibration ground clutter suppression method for an active microwave remote sensing satellite, characterized in that, The method comprises the following steps: Step S1, constructing a Doppler term in a slow time domain by using the difference between the ground speed of the active microwave remote sensing satellite and the ground speed of the space-based calibration star; Step S11, setting a linear frequency modulation (LFM) signal transmitted by the radar The expression is: （1）； wherein is an exponential function; is fast time; is pulse duration; is the imaginary unit; is the carrier frequency; is the frequency modulation rate; Step S12, echo signal from ground clutter In The expression for the time instant is: （2）； wherein, is a slow time; is an amplitude of ground clutter; is a distance of the active microwave remote sensing satellite from the ground; is a speed of light; is a wavelength; is a speed of the active microwave remote sensing satellite from the ground; Step S13, echo signal forwarded by the calibration star At The expression of the time instant is: （3）； wherein, is the echo amplitude of the calibration star; is the distance of the active microwave remote sensing satellite from the calibration star; is the velocity of the active microwave remote sensing satellite from the calibration star; Step S14, the overall received signal of the active microwave remote sensing satellite is written as a form containing ground clutter, the signal of the space-based calibration star, and noise, which is expressed at the time of the expression : （4）； wherein is a Gaussian white noise; The overall received signal of the active microwave remote sensing satellite is written as in matrix form, denoted as ; where, is the number of slow-time samples, is the number of fast-time samples; Step S2, forming different manifold space matrices based on the Doppler term in the slow time domain; Step S21, obtaining the distance and velocity of the active microwave remote sensing satellite relative to the ground according to the track information and the velocity of the active microwave remote sensing satellite relative to the calibration satellite and the velocity of the active microwave remote sensing satellite relative to the calibration satellite Step S22, forming a vector containing , a Doppler term , as follows: （5）； Let , removing the Doppler term containing the slow time , the rest is written as a matrix of dimension , as follows: （6）； （7）； wherein, is the term from the ground containing only fast-time echo information; is the term from the calibration star containing only fast-time echo information; is the sampling of the fast-time echo from the ground at each time instant; is the sampling of the fast-time echo from the calibration star at each time instant; By matrix multiplication, we get ; where is a noise matrix; Step S23, according to , forming a manifold space matrix with as follows: （8）； in, For the reason The column vectors spanned by the manifold space matrix containing ground echo information; For the reason The column vectors spanned by the manifold space matrix containing the calibration star echo information; Step S3, constructing a subspace by using the manifold matrix, performing subspace projection, and recovering the expected signal, and the specific process is as follows: The received signal is projected to the manifold space matrix, because the velocity of the active microwave remote sensing satellite relative to the ground is inconsistent with the velocity of the active microwave remote sensing satellite relative to the calibration star, so With The complex exponential term on the right is not related to, the inner product is zero, so: （9）； According to the above formula, the following is further obtained: （10）； Since the noise is uncorrelated with the signal, i.e. the estimated echo signal from the calibration star is obtained as follows: （11）。

Images