阅读说明：本技术 误差存在情况下基于真实信号子空间的目标检测方法 (Target detection method based on real signal subspace under existence of error ) 是由 吴建新 赵阳 沈鹏 李常先 于 2019-08-14 设计创作，主要内容包括：本发明属于雷达技术领域,具体涉及一种误差存在情况下基于真实信号子空间的目标检测方法,获取回波数据；根据回波数据得到主瓣杂波所占据的多个多普勒通道和主瓣杂波所占据的多个多普勒通道的中心频率；根据主瓣杂波所占据的多个多普勒通道中心频率得到多个对应空间导向矢量；根据回波数据的主杂波区得到若干滤波后输出值；根据得到多个最大输出功率对应距离门的滤波后输出值；根据若干滤波后输出值和空间导向矢量得到杂波估计的目标空域导向矢量矩阵；对杂波估计的目标空域导向矢量矩阵进行奇异值分解得到考虑误差的信号子空间；根据考虑误差的信号子空间构建子空间检测器。本发明的检测器的检测性能提高、检测概率更高。(The invention belongs to the technical field of radars, and particularly relates to a target detection method based on a real signal subspace under the condition of error, which is used for acquiring echo data; obtaining a plurality of Doppler channels occupied by the mainlobe clutter and the central frequencies of the Doppler channels occupied by the mainlobe clutter according to the echo data; obtaining a plurality of corresponding space guiding vectors according to the central frequencies of a plurality of Doppler channels occupied by the mainlobe clutter; obtaining a plurality of filtered output values according to the main clutter area of the echo data; obtaining filtered output values of the range gates corresponding to the maximum output powers; obtaining a target airspace guide vector matrix of clutter estimation according to the output values after filtering and the space guide vector; performing singular value decomposition on a target space domain guide vector matrix of clutter estimation to obtain a signal subspace considering errors; the subspace detector is constructed from the error-considered signal subspace. The detector of the invention has improved detection performance and higher detection probability.)

1. A target detection method based on a real signal subspace under the condition of existence of errors is characterized by comprising the following steps:

acquiring echo data;

obtaining a plurality of Doppler channels occupied by the mainlobe clutter and the central frequencies of the Doppler channels occupied by the mainlobe clutter according to the echo data;

obtaining a plurality of corresponding space frequencies according to a plurality of Doppler channel center frequencies occupied by the mainlobe clutter, and calculating a plurality of corresponding space guide vectors according to the plurality of space frequencies;

performing Doppler filtering processing on clutter data received by each range gate of a plurality of Doppler channels occupied by a main clutter area of the echo data to obtain a plurality of filtered output values;

performing inner product on the plurality of filtered output values and the corresponding space steering vectors to obtain a plurality of filtered output values of the range gates corresponding to the maximum output power;

obtaining a plurality of corresponding vector cone angle cosines according to the plurality of filtered output values and the space guide vector, screening the filtered output values of the range gate corresponding to the maximum output power according to the vector cone angle cosines to obtain a plurality of Doppler channel filtered output values occupied by a main clutter region of the filtered echo data, and obtaining a target airspace guide vector matrix of clutter estimation according to the plurality of Doppler channel filtered output values occupied by the main clutter region of the filtered echo data;

performing singular value decomposition on a target space domain guide vector matrix of clutter estimation to obtain a signal subspace considering errors;

the subspace detector is constructed from the error-considered signal subspace.

2. The method of real signal subspace-based target detection in the presence of errors according to claim 1, wherein deriving from said echo data a plurality of doppler channels occupied by mainlobe clutter and a plurality of doppler channel center frequencies occupied by said mainlobe clutter comprises:

obtaining a plurality of Doppler channel bandwidths occupied by a main clutter region of the echo data according to the echo data;

obtaining a plurality of normalized Doppler channel center frequencies occupied by the mainlobe clutter according to a plurality of Doppler channel bandwidths occupied by the mainlobe clutter;

and calculating the corresponding Doppler channel where the mainlobe clutter is located according to the central frequencies of the plurality of normalized Doppler channels occupied by the lobe clutter.

3. The method of claim 2, wherein obtaining a plurality of doppler channel bandwidths occupied by a clutter region of the echo data from the echo data comprises:

(a) the number of the equidistant linear array elements is N, the array element spacing is d, and the main lobe beam space cone angle is theta₀3dB beam width of theta_A＝λ/(Nd sinθ₀) Wherein Nd sin θ₀The range of spatial cone angle variation within the mainlobe beam is denoted as θ for the effective aperture length_i∈[θ₀-θ_A/2,θ₀+θ_A/2]The doppler channel bandwidth in this range is:

(b) bringing in

wherein v is the aircraft flight speed and λ is the wavelength;

(c) calculating the normalized Doppler frequency of the clutter in the range of the mainlobe:

(d) radar emitting M pulses, the resolution of the M Doppler filters being

the beam space cone angle at the center of the main lobe is theta₀Normalized Doppler frequency of

4. The method of claim 1, wherein obtaining a plurality of corresponding spatial frequencies from a plurality of doppler channel center frequencies occupied by the mainlobe clutter comprises:

spatial frequency is expressed as

5. The method of claim 4, wherein computing a plurality of corresponding spatial steering vectors from the plurality of spatial frequencies comprises:

order filter

p_imis the doppler domain filter output of the ith spur block,

6. The method of claim 1, wherein the cone angle cosine is calculated as follows:

7. The method of claim 1, wherein constructing a subspace detector from a signal subspace that accounts for errors comprises:

wherein the content of the first and second substances,

Technical Field

The invention belongs to the technical field of radars, and particularly relates to a target detection method based on a real signal subspace under the condition of error.

Background

In the array signal processing, the optimal weighting processing is carried out on the received signal vector by using the guide vector of the expected signal to obtain the filtering output, namely the guide vector of the expected signal is matched with the received signal vector, and in the conventional adaptive matching filtering processing, the space domain guide vector of the expected signal is the space domain guide vector at the center of the main lobe beam. Therefore, when the angle of the target deviates from the center of the main lobe beam, the steering vector of the target is mismatched with the preset steering vector of the center of the main lobe beam, resulting in a loss of filtering output. Further, in a matched filter detector using a signal subspace spanned by only one-dimensional guide vectors, signal information is easily lost due to guide vector mismatch.

For better detection of the target in mismatch conditions it is necessary to use a high-dimensional signal subspace detector instead of the original one-dimensional signal subspace detector. The high-dimensional signal subspace detector can accumulate signal energy in multiple dimensions, and output loss of a matched filter is reduced. In order to obtain a high-dimensional signal subspace, forming the signal subspace by using a plurality of possible steering vectors of the target direction; and decomposing the covariance matrix of the unit to be detected to obtain a signal subspace.

In practical situations, due to the existence of factors such as amplitude-phase errors of the array elements, coupling effects between the array elements, position deviations of the array elements and the like, a real signal subspace often has a certain deviation from a signal subspace under the condition that the errors are not considered. If a detector without error consideration is still used, the detector cannot effectively accumulate signal energy, resulting in filter output power loss and affecting the detection performance of the detector.

Disclosure of Invention

In order to solve the above problems in the prior art, the present invention provides a target detection method based on a real signal subspace in the presence of errors. The technical problem to be solved by the invention is realized by the following technical scheme:

the target detection method based on the real signal subspace under the condition of the existence of errors comprises the following steps:

acquiring echo data;

obtaining a plurality of Doppler channels occupied by the mainlobe clutter and the central frequencies of the Doppler channels occupied by the mainlobe clutter according to the echo data;

obtaining a plurality of corresponding space frequencies according to a plurality of Doppler channel center frequencies occupied by the mainlobe clutter, and calculating a plurality of corresponding space guide vectors according to the plurality of space frequencies;

performing Doppler filtering processing on clutter data received by each range gate of a plurality of Doppler channels occupied by a main clutter area of the echo data to obtain a plurality of filtered output values;

performing inner product on the plurality of filtered output values and the corresponding space steering vectors to obtain a plurality of filtered output values of the range gates corresponding to the maximum output power;

obtaining a plurality of corresponding vector cone angle cosines according to the plurality of filtered output values and the space guide vector, screening the filtered output values of the range gate corresponding to the maximum output power according to the vector cone angle cosines to obtain a plurality of Doppler channel filtered output values occupied by a main clutter region of the filtered echo data, and obtaining a target airspace guide vector matrix of clutter estimation according to the plurality of Doppler channel filtered output values occupied by the main clutter region of the filtered echo data;

performing singular value decomposition on a target space domain guide vector matrix of clutter estimation to obtain a signal subspace considering errors;

the subspace detector is constructed from the error-considered signal subspace.

According to the invention, the signal subspace considering the error is obtained, and the sub-control detector obtained through the signal subspace has higher detection probability under the condition of a certain SNR, so that the detection performance of the detector is improved.

In one embodiment of the present invention, obtaining a plurality of doppler channels occupied by mainlobe clutter and a plurality of doppler channel center frequencies occupied by the mainlobe clutter from the echo data comprises:

obtaining a plurality of Doppler channel bandwidths occupied by a main clutter region of the echo data according to the echo data;

obtaining a plurality of normalized Doppler channel center frequencies occupied by the mainlobe clutter according to a plurality of Doppler channel bandwidths occupied by the mainlobe clutter;

and calculating the corresponding Doppler channel where the mainlobe clutter is located according to the central frequencies of the plurality of normalized Doppler channels occupied by the lobe clutter.

In one embodiment of the present invention, obtaining a plurality of doppler channel bandwidths occupied by a main clutter region of the echo data from the echo data comprises:

(a) considering the equidistant linear array of N array elements, the distance between the array elements is d, and the space cone angle of the main lobe beam is theta₀3dB beam width of theta_A＝λ/(Nd sinθ₀) Wherein Nd sin θ₀The effective aperture length, and therefore the range of spatial cone angle variation within the mainlobe beam, is denoted as θ_i∈[θ₀-θ_A/2,θ₀+θ_A/2]The doppler channel bandwidth in this range is:

(b) bringing inObtaining the bandwidth of the Doppler channel corresponding to the mainlobe clutter as follows:

wherein v is the aircraft flight speed and λ is the wavelength;

(c) calculating the normalized Doppler frequency of the clutter in the range of the mainlobe:

(d) radar emitting M pulses, the resolution of the M Doppler filters being

Impurities in the main lobeThe approximate number of doppler channels occupied by a wave is:

the beam space cone angle at the center of the main lobe is theta₀Normalized Doppler frequency of

The multiple Doppler channels occupied by the mainlobe clutter are denoted B_d0So that the Doppler channel occupied by the clutter zone calculated according to the width of the mainlobe beam is

In one embodiment of the invention, the spatial frequency is calculated as follows:

spatial frequency is expressed as

Wherein d is the array element spacing, λ is the wavelength,is the array space cone angle.

In one embodiment of the present invention, calculating a plurality of corresponding spatial steering vectors from the plurality of spatial frequencies comprises:

order filterCorresponding normalized Doppler frequency ofCorresponding to a spatial frequency of f_smSampling data x of the ith range gate of the receiver_lIs shown as

In the formula N_cThe number of clutter blocks in the clutter equidistant ring is represented,

is the normalized doppler frequency of the corresponding ith clutter block,

is the time-oriented vector of the spur block;

is the spatial frequency corresponding to the ith spur block,

for a spatially oriented vector, n is Gaussian noise, and the data pass band is

Doppler filter ofThe latter output is expressed as:

p_imis the doppler domain filter output of the ith spur block,

is the output of the noise after passing through the filter, and further obtains the corresponding space guide vector expression of the noise as

In one embodiment of the invention, the cone angle cosine is calculated as follows:

y_mlthe data of the first range gate is the output value after passing through the mth Doppler filter, wherein L is more than or equal to 0 and less than or equal to L₀，L₀Is the total number of distance gates; m is more than 0; a is_mIs a spatial steering vector.

In one embodiment of the invention, constructing a subspace detector from a signal subspace that accounts for errors comprises:

wherein the content of the first and second substances,

x_pis H₁And H₀In both cases, the received data, U is the signal subspace,

is a noise covariance matrix.

The invention has the beneficial effects that:

according to the invention, the signal subspace considering the error is obtained, and the sub-control detector obtained through the signal subspace has higher detection probability under the condition of a certain SNR, so that the detection performance of the detector is improved.

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

Drawings

Fig. 1 is a block flow diagram of a target detection method based on a real signal subspace in the presence of an error according to an embodiment of the present invention;

fig. 2 is a block flow diagram of a target detection method based on a real signal subspace in the presence of another error according to an embodiment of the present invention;

fig. 3 is a relationship between an angle estimation error and an SNR change when a main lobe beam is positively swept and a magnitude-phase error is 10% in a target detection method based on a real signal subspace in the presence of an error according to an embodiment of the present invention;

fig. 4 is a relationship between an angle estimation error and an SNR change when a main lobe center is swept and a magnitude-phase error is 10% in a target detection method based on a real signal subspace in the presence of an error according to an embodiment of the present invention;

fig. 5 is a relationship between an angle estimation mean square error and a target incident angle change when a main lobe beam is positively swept and a magnitude-phase error is 10% in the target detection method based on a real signal subspace in the presence of an error according to an embodiment of the present invention;

fig. 6 is a relationship between an angle estimation mean square error and a target incident angle change when a main lobe beam is swept and a magnitude-phase error is 10% in the target detection method based on a real signal subspace in the presence of an error according to an embodiment of the present invention.

Detailed Description

The present invention will be described in further detail with reference to specific examples, but the embodiments of the present invention are not limited thereto.

Referring to fig. 1, fig. 1 is a block flow diagram of a target detection method based on a real signal subspace in the presence of an error according to an embodiment of the present invention, including:

acquiring echo data;

obtaining a plurality of Doppler channels occupied by the mainlobe clutter and the central frequencies of the Doppler channels occupied by the mainlobe clutter according to the echo data;

obtaining a plurality of corresponding space frequencies according to a plurality of Doppler channel center frequencies occupied by the mainlobe clutter, and calculating a plurality of corresponding space guide vectors according to the plurality of space frequencies;

performing Doppler filtering processing on clutter data received by each range gate of a plurality of Doppler channels occupied by a main clutter area of the echo data to obtain a plurality of filtered output values;

performing inner product on the plurality of filtered output values and the corresponding space steering vectors to obtain a plurality of filtered output values of the range gates corresponding to the maximum output power;

obtaining a plurality of corresponding vector cone angle cosines according to the plurality of filtered output values and the space guide vector, screening the filtered output values of the range gate corresponding to the maximum output power according to the vector cone angle cosines to obtain a plurality of Doppler channel filtered output values occupied by a main clutter region of the filtered echo data, and obtaining a target airspace guide vector matrix of clutter estimation according to the plurality of Doppler channel filtered output values occupied by the main clutter region of the filtered echo data;

performing singular value decomposition on a target space domain guide vector matrix of clutter estimation to obtain a signal subspace considering errors;

the subspace detector is constructed from the error-considered signal subspace.

In one embodiment of the present invention, obtaining a plurality of doppler channels occupied by mainlobe clutter and a plurality of doppler channel center frequencies occupied by the mainlobe clutter from the echo data comprises:

obtaining a plurality of Doppler channel bandwidths occupied by a main clutter region of the echo data according to the echo data;

obtaining a plurality of normalized Doppler channel center frequencies occupied by the mainlobe clutter according to a plurality of Doppler channel bandwidths occupied by the mainlobe clutter;

and calculating the corresponding Doppler channel where the mainlobe clutter is located according to the central frequencies of the plurality of normalized Doppler channels occupied by the lobe clutter.

In an embodiment of the present invention, please refer to fig. 2, fig. 2 is a block flow diagram of a target detection method based on a real signal subspace in the presence of another error according to an embodiment of the present invention, in which a plurality of doppler channel bandwidths occupied by a main clutter region of echo data are obtained according to the echo data, including:

(a) considering the equidistant linear array of N array elements, the distance between the array elements is d, and the space cone angle of the main lobe beam is theta₀3dB beam width of theta_A＝λ/(Nd sinθ₀) Wherein Nd sin θ₀The effective aperture length, and therefore the range of spatial cone angle variation within the mainlobe beam, is denoted as θ_i∈[θ₀-θ_A/2,θ₀+θ_A/2]The doppler channel bandwidth in this range is:

(b) bringing in

Obtaining the bandwidth of the Doppler channel corresponding to the mainlobe clutter as follows:

wherein v is the aircraft flight speed and λ is the wavelength;

(c) calculating the normalized Doppler frequency of the clutter in the range of the mainlobe:

(d) radar emitting M pulses, the resolution of the M Doppler filters being

The approximate number of doppler channels occupied by clutter in the mainlobe is:

the beam space cone angle at the center of the main lobe is theta₀Normalized Doppler frequency ofThe multiple Doppler channels occupied by the mainlobe clutter are denoted B_d0So that the Doppler channel occupied by the clutter zone calculated according to the width of the mainlobe beam is

In one embodiment of the invention, the spatial frequency is calculated as follows:

spatial frequency is expressed as

Wherein d is an arrayThe cell spacing, λ is the wavelength,

is the array space cone angle.

Further, the normalized Doppler frequency is

The relationship between them is shown as follows:

in one embodiment of the present invention, calculating a plurality of corresponding spatial steering vectors from the plurality of spatial frequencies comprises:

order filterCorresponding normalized Doppler frequency ofCorresponding to a spatial frequency of f_smSampling data x of the ith range gate of the receiver_lIs shown as

In the formula N_cThe number of clutter blocks in the clutter equidistant ring is represented,is the normalized doppler frequency of the corresponding ith clutter block,

is the time-oriented vector of the spur block;

is the spatial frequency corresponding to the ith spur block,for a spatially oriented vector, n is Gaussian noise, and the data pass band isDoppler filter of

The latter output is expressed as:

p_imis the doppler domain filter output of the ith spur block,

is the output of the noise after passing through the filter, and further obtains the corresponding space guide vector expression of the noise as

Further, according to the output value y_mlAnd a space guide vector a_mInner product formula gamma_l＝a_m ^Ηy_mlSelecting L with the maximum output power₀Filtered output value y of range gate_mThe selected range gate is represented as

The selected data is expressed as

In one embodiment of the invention, the cone angle cosine is calculated as follows:

y_mlthe data of the first range gate is the output value after passing through the mth Doppler filter, wherein L is more than or equal to 0 and less than or equal to L₀，L₀Is the total number of distance gates; m is more than 0; a is_mIs a spatial steering vector.

Further, the cone angle cosine μ is selected_lDistance door l corresponding to maximum time_maxOutput of (2)

Represents the space cone angle f_smCorresponding error-considered steering vectors.

Further, a target space-domain steering vector matrix of clutter estimation is obtained

Further, singular value decomposition is carried out on a target airspace guide vector matrix of clutter estimation to obtain a signal subspace considering errors, and singular value decomposition of the guide vector matrix is carried out to obtain ULambda V^ΗTaking the signal feature subspace as U_eig＝U(:,1:2)。

In one embodiment of the invention, constructing a subspace detector from a signal subspace that accounts for errors comprises:

wherein the content of the first and second substances,x_pis H₁And H₀In both cases, the received data, U is the signal subspace,

is a noise covariance matrix.

The effect of the invention can be further illustrated by the following simulation experiment:

1. simulation conditions are as follows:

the environment of the simulation experiment of the invention is as follows: MATLAB2017b, Intel (R) Xeon (R) CPU2.20GHz, Windows7 professional edition

2. Simulation parameters:

consider an equidistant linear array, wave, with 8 elementsThe length is 0.4 m, and the spacing between array elements is half wavelength. The speed of the carrier is 100m/s, and the repetition frequency is 2000 Hz. To obtain more mainlobe clutter data, the number of pulses is set to 256, and the mainlobe beam is directed in the normal direction. The doppler bandwidth of the mainlobe clutter within the 3dB beamwidth is calculated as 125Hz, the resolution of the doppler filter is 7.8125, and the calculated mainlobe clutter passes through 16 doppler filters. Thus, a succession of 11 doppler filters at the center of the doppler channel of the main lobe are selected. The Doppler channel at the center of the mainlobe is the 129 th Doppler channel, so that the selected Doppler channel of the mainlobe clutter is the 124 th to 134 th Doppler channels. The mainlobe Doppler channel is 184 when the mainlobe is swept by 60 degrees, and the selected mainlobe clutter Doppler channel is 179-189. Obtaining data y of mainlobe clutter according to Doppler filter_mlAnd then, obtaining an estimation subspace through singular value decomposition.

3. Simulation results and analysis:

simulation 1: referring to fig. 3, fig. 3 shows a relationship between a change of an angle estimation error and an SNR when a mainlobe beam is positively swept and a phase-amplitude error is 10% in a target detection method based on a real signal subspace in the presence of an error according to an embodiment of the present invention, where a conventional sum and difference beam angle measurement method and a feature subspace angle measurement method are adopted, and a feature subspace includes a feature subspace without an error considered and a feature subspace of clutter estimation. Because the feature subspace estimated by using the real clutter data can better estimate the target guide vector with errors, the target guide vector has better performance in detection. Similar to detection, the characteristic subspace of clutter estimation is used for grid search angle measurement, and the angle measurement performance of the clutter estimation subspace method is verified in a simulation mode. As can be seen from fig. 3, when the SNR (signal-to-noise ratio) is low, the detection performance of each subspace is poor, and the angle measurement error is large. And as the SNR is increased, the angle measurement error of each subspace is smaller and smaller, and the subspace approaches to be stable.

Simulation 2: referring to fig. 4, fig. 4 shows a relationship between angle estimation errors and SNR when the center of the main lobe is swept and the amplitude-phase error is 10% in the target detection method based on the real signal subspace under the existence of errors, and it can be seen from the figure that the accuracy of each angle measurement method is higher and higher as the SNR increases. When the main lobe is swept, the conventional and poor beam angle measurement performance is greatly reduced, and the real subspace angle measurement performance is improved to the maximum extent as the SNR is increased.

Simulation 3: referring to fig. 5, fig. 5 is a relationship that an angle estimation mean square error changes with a target incident angle when a main lobe beam is positively swept and a magnitude-phase error is 10% in a target detection method based on a real signal subspace under the existence of an error, and it can be known from the figure that an angle estimation error of a conventional sum-difference beam angle measurement method increases with the increase of the error; the clutter estimation feature subspace is estimated from the clutter, the error is relatively small, and the combination of the clutter estimation feature subspace and the maximum likelihood angle measurement method has the best angle estimation performance.

And (4) simulation: referring to fig. 6, fig. 6 shows a relationship between a mainlobe beam deflection and an angle estimation mean square error changing with a target incident angle when a phase error is 10% in a target detection method based on a real signal subspace under the presence of an error according to an embodiment of the present invention.

The experimental result shows that the signal subspace obtained by using the estimation of the main lobe clutter data has better detection performance compared with the signal subspace without considering the error under the condition of the array amplitude-phase error and the array element position deviation. By performing angle measurement analysis on the target data by using each subspace, the angle measurement performance of the clutter estimated signal subspace can be seen to be better.

The foregoing is a more detailed description of the invention in connection with specific preferred embodiments and it is not intended that the invention be limited to these specific details. For those skilled in the art to which the invention pertains, several simple deductions or substitutions can be made without departing from the spirit of the invention, and all shall be considered as belonging to the protection scope of the invention.