Non-stationary clutter suppression method for end-fire array airborne radar based on mixed domain dimension reduction

Abstract

The application discloses a mixed domain dimension reduction based end-on array airborne radar non-stationary clutter suppression method and belongs to the field of radar signal processing, and comprises the following steps: digitized echo data is combined into multiple row space channels through a subarray in a horizontal row direction; the subarray is converted into a beam domain through a discrete Fourier transform in a horizontal column direction, and a main beam and several adjacent auxiliary beams are selected; column beams and row space channels are mixed to obtain space domain data; space-time receiving data is obtained through space domain conversion of the space domain data; a space-time adaptive processing (STAP) method is used for clutter suppression processing, clutter suppression is completed; constant false alarm rate (CFAR) processing is performed on the data after the clutter suppression, and finally, a moving target is detected. The application is a mixed domain dimension reduction three-dimensional clutter decoupling method with low system degrees of freedom, so that effective suppression of non-stationary clutter can be realized.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of radar signal processing, and more particularly, to an end-fire array airborne radar non-stationary clutter suppression method based on mixed domain dimension reduction. BACKGROUND

[0002] Modern war puts forward higher and higher requirements for the maneuverability and target detection capability of airborne early warning radar. End-fire array antenna is very suitable for solving the front and rear blind area problem of airborne radar because of its low profile and the maximum radiation direction pointing to the array axial direction. In this regard, foreign countries have relatively mature technology and have been applied in practice. The most representative one is the E-737 early warning aircraft developed by Boeing Company of the United States. Similar to the traditional side-fire array airborne radar, the end-fire array airborne early warning radar also faces the influence of strong ground / sea clutter when working in the downward direction. Therefore, the first problem to be solved in the practical application of end-fire array airborne radar is the clutter suppression. Space-time adaptive processing (STAP) is the key technology for airborne radar system to suppress clutter. In the past few decades, domestic and foreign scholars have carried out a large number of researches on the STAP method of side-fire array airborne radar, and some classical STAP methods have been applied in practice. However, since the end-fire array is usually installed horizontally on the platform, and its main beam direction is parallel to the array surface, the clutter of end-fire array airborne radar presents different distribution characteristics from the side-fire array airborne radar. Specifically, the clutter of end-fire array airborne radar presents a three-dimensional coupling relationship in the row space frequency, column space frequency and Doppler frequency, which leads to significant differences in the clutter characteristics of different ambiguous distances, i.e. the non-stationarity of clutter. Therefore, if the traditional STAP method is used to suppress the non-stationary clutter of end-fire array, the performance will be significantly reduced.

[0003] The existing methods for suppressing the non-stationary clutter of side-fire array airborne radar mainly include the following two categories:

[0004] 1) Clutter compensation method, including Doppler compensation method, angle-Doppler compensation method, adaptive angle-Doppler compensation method and clutter spectrum registration based compensation method. This kind of method generally takes the unit to be detected as the reference, realizes the frequency shift of clutter spectrum in the frequency domain through compensation phase, and reduces the influence of clutter non-stationarity on clutter suppression. However, the clutter compensation method is only suitable for the problem that the clutter distribution is inconsistent between different training samples. In the case of distance ambiguity, this kind of algorithm cannot simultaneously correct the clutter spectrum of different ambiguous distances to be consistent, which leads to a sharp decline in performance.

[0005] 2) The non-stationary clutter suppression method combined with the elevation dimension of the side-looking array, mainly including the elevation pre-filtering cascaded azimuth-Doppler two-dimensional adaptive processing method and the azimuth-elevation-Doppler three-dimensional STAP and the corresponding dimension reduction method. This kind of method mainly utilizes the difference of the clutter in the elevation dimension to suppress the non-stationary clutter. However, the end-on array is usually horizontally placed on the moving platform, and there is no elevation spatial freedom, so this kind of method cannot be directly applied to the clutter suppression processing of the end-on array airborne early warning radar.

[0006] The existing research on the end-on array airborne radar clutter suppression mainly focuses on the two-dimensional adaptive processing in the azimuth-Doppler domain. Actually, the end-on array airborne radar clutter is three-dimensionally coupled in the row spatial frequency domain-column spatial frequency domain-Doppler frequency domain. Therefore, if the method of jointly decoupling the row spatial domain, the column spatial domain and the Doppler domain is considered, the influence of the non-stationary clutter can be effectively avoided, and the clutter suppression performance can be improved. However, if the clutter suppression processing is performed in the full-dimensional row, column and Doppler domain, it will inevitably lead to huge computation and uniform sample requirement, and thus it is difficult to be applied to engineering practice. SUMMARY

[0007] In order to solve the problems and defects existing in the prior art, based on the coupling law of the end-on array clutter in the row spatial frequency, column spatial frequency and Doppler three-dimensional space, the application provides a kind of end-on array airborne radar non-stationary clutter suppression method based on mixed domain dimension reduction, which is a kind of mixed domain dimension reduction three-dimensional clutter decoupling method with low system freedom, so as to realize the effective suppression of non-stationary clutter.

[0008] In order to achieve the above-mentioned purpose of the application, the technical scheme adopted is as follows:

[0009] An end-on array airborne radar non-stationary clutter suppression method based on mixed domain dimension reduction, the method comprises the following steps:

[0010] The digitized echo data is combined into a plurality of row spatial domain channels through the horizontal row subarray;

[0011] The horizontal column subarray is converted to the beam domain through the discrete Fourier transform, and the main beam and a plurality of adjacent auxiliary beams are selected;

[0012] The column beam and the row spatial domain channel are mixed to obtain spatial domain data;

[0013] The spatial domain data is converted to spatial-time receiving data through spatial domain conversion, and the spatial-time adaptive processing method STAP is used for clutter suppression processing, so as to complete the suppression of the clutter;

[0014] The data after completing the clutter suppression is subjected to constant false alarm CFAR processing, and finally the detection of the moving target is completed.

[0015] Preferably, the horizontal row direction synthesizes multiple row direction spatial channels by using non-weighted subarray synthesis, and uses adjacent subarrays or sliding window subarrays for the structure of the subarrays.

[0016] Further, the horizontal column direction forms adjacent multiple beams by spatial discrete Fourier transform, wherein the main beam is located at the center of the multiple beams.

[0017] Preferably, the digitized echo data is obtained by down-converting the echo signals received by the airborne planar end-fire array and then performing analog-digital conversion.

[0018] Preferably, N ch row direction spatial channels are synthesized in the direction of the row subarray of the horizontal row direction, and each row direction spatial channel contains N sub array elements, and the row direction conversion matrix is represented as:

[0019]

[0020] wherein, is the Kronecker product; is a column vector whose nth element is 1 and the rest are 0; is the spatial steering vector of the horizontal row direction subarray synthesis channel, that is,

[0021]

[0022] wherein, d r is the row subarray element spacing, λ is the radar operating wavelength, θ0 and are the main beam azimuth angle and elevation angle, respectively.

[0023] Further, the echo data is converted to the beam domain along the horizontal column direction, and the main beam and its adjacent 2M b beams are selected, and the conversion matrix is represented as:

[0024] T sc = [s sc (ω sc ), s sc (ω sc ±ω p ),...,s sc (ω sc ±M b ω p )] (9)

[0025] wherein, ω p is the adjacent beam spacing, ω sc represents the spatial column direction angular frequency corresponding to the main beam, and M represents the number of horizontal column direction array elements, Mb denotes the number of selected beams, d c is the column subarray element spacing, s sc (ω) denotes the column-wise beam steering vector, i.e.

[0026] s sc (ω) = [1, e jω ,..., e j(M-1)ω ] T (10).

[0027] Further, the column-wise beam and the row-wise spatial channel are mixed to obtain spatial data, which is specifically as follows:

[0028] The spatial conversion matrix is denoted as The converted spatial data is x k denotes the original spatial receiving data;

[0029] For the number of pulses in the coherent processing time being K, the space-time receiving data of K pulses is:

[0030]

[0031] Further, the converted space-time receiving data is processed by a space-time adaptive processing method STAP for clutter suppression, specifically, the STAP adaptive weight is obtained according to a minimum variance distortionless response criterion MVDR criterion:

[0032]

[0033] wherein, is the clutter covariance matrix after time domain dimension reduction; (·) H denotes the conjugate transpose, T t is the time domain dimension reduction matrix; E[·] denotes mathematical expectation; s T is the converted space-time steering vector, s T = T H s, is the space-time steering vector; s t , s r and s c are respectively the time domain steering vector, the spatial row-wise steering vector and the spatial column-wise steering vector.

[0034] A computer device comprises a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the processor executes the computer program to implement the steps of the method.

[0035] A computer readable storage medium, having stored thereon a computer program, which, when executed by a processor, implements the steps of the method as described above.

[0036] The beneficial effects of the present application are as follows:

[0037] The present application is based on the space-time coupling law of end-fire array clutter in the row space-frequency, column space-frequency and Doppler three-dimensional domain, and invents a row multi-channel, column multi-beam and time domain three-dimensional joint adaptive processing scheme, which effectively improves the non-stationary clutter suppression performance of the end-fire array. Compared with the traditional column space-frequency-time domain two-dimensional adaptive processing, the present application can significantly improve the non-stationary clutter suppression performance of the end-fire array; compared with the full-dimensional column space-frequency-row space-frequency-time domain adaptive processing, the present application can obtain similar non-stationary clutter suppression while significantly reducing the operation amount and the sample demand required for estimating the clutter covariance matrix, and thus is more beneficial to practical engineering application. BRIEF DESCRIPTION OF DRAWINGS

[0038] Figure 1 The present embodiment is a non-stationary clutter suppression method for end-fire array airborne radar based on mixed domain dimension reduction. DETAILED DESCRIPTION

[0039] The present application will be described in detail below in combination with the drawings and specific embodiments.

[0040] Embodiment 1

[0041] The present embodiment considers the distribution characteristics of the end-fire array clutter in the row space-frequency-Doppler domain, i.e. all the clutter components outside the main clutter, including all the near-range echoes and the sidelobe clutter components of the multiple range ambiguity echoes, are located in the row space-frequency sidelobe region, so the horizontal row can be combined into multiple subarray channels for joint Doppler adaptive processing. The main clutter component exists on the row space-frequency main lobe, but it is different in the horizontal column space-frequency, so it can be suppressed in the form of column space-frequency main beam combined with adjacent beams. This is because the array elements / channels participate in adaptive processing, which is beneficial to suppress the discrete clutter from different angles; and the beam domain participates in adaptive processing, which is more suitable for suppressing the continuously distributed main lobe clutter. Considering the above factors, the present embodiment proposes a spatial domain dimension reduction method of row subarray combined multi-channel joint column space-frequency multi-beam for clutter suppression processing, which includes the following technical steps:

[0042] As shown in Figure 1 A non-stationary clutter suppression method for end-fire array airborne radar based on mixed domain dimension reduction, the method includes the following steps:

[0043] The digitized echo data is combined into multiple row space domain channels in the horizontal row direction by subarray;

[0044] The horizontal column direction sub-arrays are converted into beam domain by discrete Fourier transform, and a main beam and several auxiliary beams adjacent to the main beam are selected;

[0045] The column direction beam and the row direction space domain channel are mixed to obtain space domain data;

[0046] The space domain data are converted to obtain space-time receiving data, and a space-time adaptive processing (STAP) method is used for clutter suppression processing to complete the suppression of the clutter;

[0047] The data after the completion of the clutter suppression are subjected to constant false alarm (CFAR) processing, and finally the detection of the moving target is completed.

[0048] In the embodiment, the horizontal row direction uses a non-weighted sub-array synthesis method to form a plurality of row direction space domain channels, and the structure of the sub-array uses an adjacent sub-array or a sliding window sub-array. In the embodiment, the digitized echo data are synthesized along the horizontal row direction, and a plurality of row direction space domain channels are synthesized for each row.

[0049] The horizontal column direction forms adjacent multiple beams by space domain discrete Fourier transform, and the main beam is located at the center of the multiple beams.

[0050] According to the minimum variance distortionless response (MVDR) criterion, the space-time echo data after the completion of the space domain conversion are subjected to STAP processing, and then the effective suppression of the non-stationary clutter of the airborne early warning radar of the end-fire array is realized. The time domain can use pulse domain data or adjacent multiple Doppler channel data after Fourier transform.

[0051] In the embodiment, the digitized echo data are obtained by down-conversion processing of the echo signals received by the airborne planar end-fire array and then analog-digital conversion.

[0052] Embodiment 2

[0053] Based on the end-fire array airborne radar non-stationary clutter suppression method based on mixed domain dimension reduction in embodiment 1, it is assumed that the horizontal row direction and the horizontal column direction array elements of the end-fire array airborne early warning radar are N and M respectively, the number of pulses in the coherent processing time is K, in the example, N = 20, M = 10, K = 16, and the following combined with Figure 1 and the example explain the whole invention detailed steps:

[0054] In the embodiment, the all-digital array end-fire array airborne early warning radar is taken as an example for illustration, and it is assumed that the angle between the aircraft motion speed and the horizontal row direction array normal is 60°.

[0055] The echo signal received by the airborne early warning radar space receiving channel is down-converted, then analog-to-digital converted, and the digitized echo data is stored in the system. The digitized space receiving data corresponding to the kth pulse can be expressed as:

[0056]

[0057] wherein x m,n represents the data received by the element located at the mth row and the nth column. It is reconstructed into a column vector form:

[0058] x k =[x 1,1 x 2,1 … x M,1 x 1,2 x 2,2 … x M,2 … x 1,N … x M,N ] T (2)

[0059] The echo data is subarray synthesized along the horizontal row direction, considering that N ch channels are synthesized in the row subarray direction, each channel containing N sub elements, in this example N ch = 4, N sub = 17, and the row conversion matrix can be expressed as:

[0060]

[0061] wherein, is the Kronecker product; is a column vector whose nth element is 1 and the rest are 0; is the space domain steering vector of the horizontal row subarray synthesis channel, that is:

[0062]

[0063] wherein d r is the row subarray element spacing, λ is the radar operating wavelength, θ0 and are the main beam azimuth angle and the elevation angle, respectively.

[0064] The echo data is converted to the beam domain along the horizontal column direction, and the main beam and its adjacent 2M b beams are selected, and the conversion matrix is expressed as:

[0065] T sc =[s sc (ω sc ),ssc (ω sc ±ω p ),...,s sc (ω sc ±M b ω p )] (5)

[0066] where ω p is the adjacent beam interval, ω sc represents the spatial column steering angle frequency corresponding to the main beam, and M represents the number of horizontal column array elements, M b represents the number of selected beams, d c is the column subarray element spacing, s sc (ω) represents the column beam conversion vector, i.e.

[0067] s sc (ω)=[1,e jω ,...,e j(M-1)ω ] T (6)。

[0068] In this embodiment, the column beam and the row spatial channel are mixed to obtain spatial data, which is as follows:

[0069] The spatial conversion matrix is represented as The converted spatial data is x k represents the original spatial receiving data;

[0070] For the number of pulses K within the coherent processing time, the space-time receiving data of K pulses is:

[0071]

[0072] In this embodiment, the converted space-time receiving data is processed by the space-time adaptive processing method STAP for clutter suppression. Specifically, the STAP adaptive weight is obtained according to the minimum variance distortionless response criterion MVDR criterion:

[0073]

[0074] wherein, is the clutter covariance matrix after time domain dimension reduction; (·) H represents the conjugate transpose, T t is the time domain dimension reduction matrix; E[·] represents the mathematical expectation; s T is the converted space-time steering vector, s T = T H s, is the space-time steering vector; s t, s r and s c The time domain steering vector, the space domain row steering vector and the space domain column steering vector can be respectively expressed as

[0075]

[0076]

[0077]

[0078] Herein f r is the pulse repetition frequency, and v is the platform flight speed.

[0079] The adaptive processed data is subjected to CFAR detection to complete the detection and processing of the moving target of the end-on array airborne early warning radar.

[0080] The embodiment solves the following technical problems:

[0081] 1) The end-on array row direction (end-on direction) forms multiple channels to participate in adaptive processing, and suppresses all row space frequency sidelobe clutter including distance ambiguity non-stationary.

[0082] 2) The end-on array direction (side-on direction) forms adjacent multiple beams to participate in adaptive processing, and suppresses the remaining row / column space frequency continuous main lobe clutter.

[0083] 3) The row direction multiple channels and the column direction adjacent multiple beams jointly perform space-time adaptive processing in time domain, and effectively suppress non-stationary clutter.

[0084] Embodiment 3

[0085] A computer device comprises a memory, a processor, and a computer program stored in the memory and executable on the processor, and the processor executes the computer program to realize the steps of the method of embodiment 1 or embodiment 2.

[0086] Embodiment 4

[0087] A computer readable storage medium, which stores a computer program, and the computer program is executed by a processor to realize the steps of the method of embodiment 1 or embodiment 2.

[0088] Obviously, the above embodiments of the present application are only examples for clearly illustrating the present application, and are not intended to limit the implementation modes of the present application. Any modification, equivalent replacement and improvement within the spirit and principle of the present application should be included in the protection scope of the claims of the present application.

Claims

1. A method for non-stationary clutter suppression of an airborne radar with end-fire array based on mixed-domain dimensionality reduction, characterized in that: The method comprises the following steps: The digitized echo data is combined into a plurality of row space channels in horizontal row direction by subarray; The column direction subarray is converted into beam domain by discrete Fourier transform, and a main beam and several adjacent auxiliary beams are selected; The column direction beam and the row direction space channel are mixed to obtain space domain data; The space domain data is converted into space-time receiving data by space domain conversion, and clutter suppression processing is performed by space-time adaptive processing method STAP to complete clutter suppression; CFAR processing is performed on the data after clutter suppression, and finally the detection of the moving target is completed.

2. The hybrid-domain dimensionality reduction based non-stationary clutter suppression method for an end-fire array airborne radar according to claim 1, characterized in that: The horizontal row direction adopts a non-weighted subarray synthesis method to form a plurality of row direction space channels, and an adjacent subarray or a sliding window subarray is used for the structure of the subarray.

3. The hybrid domain dimensionality reduction based non-stationary clutter suppression method for an end-fire array airborne radar according to claim 1, characterized in that: The horizontal column direction forms adjacent multi-beams by space domain discrete Fourier transform, and the main beam is located at the center of the multi-beam.

4. The hybrid domain dimensionality reduction based non-stationary clutter suppression method for an end-fire array airborne radar according to claim 1, characterized in that: The digitized echo data is obtained by down-conversion processing of the echo signal received by the airborne planar end-on array, and then analog-digital conversion.

5. The hybrid-domain dimensionality reduction based non-stationary clutter suppression method for an end-fire array airborne radar according to claim 1, characterized in that: Consider synthesizing N ch horizontal row-wise subarray direction, each row-wise spatial channel containing N sub array elements, the row-wise transformation matrix is represented as: wherein is the Kronecker product; is a column vector with the nth element being 1 and the rest being 0; is the spatial steering vector of the horizontal row-wise sub-channel synthesis channel, i.e.: where d r is the row subarray element spacing, λ is the radar operating wavelength, θ0and are the main beam azimuth and elevation angles, respectively.

6. The hybrid-domain dimensionality-reduced monopole array airborne radar non-stationary clutter suppression method according to claim 5, characterized in that: The echo data is converted to beam domain along the horizontal column, and the main beam and its adjacent 2M b beam are selected, and the conversion matrix is represented as: T sc = [s sc (ω sc ), s sc (ω sc ± ω p ),..., s sc (ω sc ± M b ω p )] (3) where ω p is the adjacent beam interval, ω sc denotes the spatial column steering angle frequency corresponding to the main beam, and M denotes the number of horizontal column steering elements, M b denotes the number of selected beams, d c is the column subarray element spacing, s sc (ω) denotes the column steering beam conversion vector, i.e. s sc (ω) = [1, e jω ,...,e j(M-1)ω ] T (4).

7. The hybrid-domain dimensionality-reduced monopole array airborne radar non-stationary clutter suppression method according to claim 6, characterized in that: The column direction beam and the row direction space channel are mixed to obtain space domain data, and the specific steps are as follows: The spatial conversion matrix is represented as The converted spatial data is x k denotes the spatial original received data; For K pulses in the coherent processing time, the space-time receiving data of K pulses is:

8. The hybrid-domain dimensionality-reduced end-fire array on-board radar non-stationary clutter suppression method according to claim 7, characterized in that: The converted space-time receiving data is processed by space-time adaptive processing method STAP for clutter suppression, and specifically, the STAP adaptive weight is obtained according to the minimum variance distortionless response criterion MVDR criterion: wherein is the clutter covariance matrix after dimensionality reduction in time domain; H denotes the conjugate transpose, T t is the dimensionality reduction matrix in time domain; E[·] denotes the mathematical expectation; s T is the transformed space-time steering vector, s T = T H s, is the space-time steering vector; s t , s r and s c are the time domain steering vector, the spatial row steering vector and the spatial column steering vector, respectively.

9. A computer device comprising a memory, a processor, and a computer program stored on the memory and executable on the processor, characterized in that: The processor executes the computer program to realize the steps of the method of any one of claims 1-8.

10. A computer readable storage medium having stored thereon a computer program, characterized in that: The computer program is executed by the processor to realize the steps of the method of any one of claims 1-8.

Images