CN113534149A - Distributed radar imaging method based on MIAA - Google Patents

Distributed radar imaging method based on MIAA Download PDF

Info

Publication number
CN113534149A
CN113534149A CN202110627285.1A CN202110627285A CN113534149A CN 113534149 A CN113534149 A CN 113534149A CN 202110627285 A CN202110627285 A CN 202110627285A CN 113534149 A CN113534149 A CN 113534149A
Authority
CN
China
Prior art keywords
data
sampled
radar
sampled data
sampling
Prior art date
Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
Pending
Application number
CN202110627285.1A
Other languages
Chinese (zh)
Inventor
戴奉周
周璇
韩彤
Current Assignee (The listed assignees may be inaccurate. Google has not performed a legal analysis and makes no representation or warranty as to the accuracy of the list.)
Xidian University
Original Assignee
Xidian University
Priority date (The priority date is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the date listed.)
Filing date
Publication date
Application filed by Xidian University filed Critical Xidian University
Priority to CN202110627285.1A priority Critical patent/CN113534149A/en
Publication of CN113534149A publication Critical patent/CN113534149A/en
Pending legal-status Critical Current

Links

Images

Classifications

    • G—PHYSICS
    • G01—MEASURING; TESTING
    • G01S—RADIO DIRECTION-FINDING; RADIO NAVIGATION; DETERMINING DISTANCE OR VELOCITY BY USE OF RADIO WAVES; LOCATING OR PRESENCE-DETECTING BY USE OF THE REFLECTION OR RERADIATION OF RADIO WAVES; ANALOGOUS ARRANGEMENTS USING OTHER WAVES
    • G01S13/00—Systems using the reflection or reradiation of radio waves, e.g. radar systems; Analogous systems using reflection or reradiation of waves whose nature or wavelength is irrelevant or unspecified
    • G01S13/88—Radar or analogous systems specially adapted for specific applications
    • G01S13/89—Radar or analogous systems specially adapted for specific applications for mapping or imaging
    • G01S13/90—Radar or analogous systems specially adapted for specific applications for mapping or imaging using synthetic aperture techniques, e.g. synthetic aperture radar [SAR] techniques
    • G01S13/9021—SAR image post-processing techniques
    • G—PHYSICS
    • G01—MEASURING; TESTING
    • G01S—RADIO DIRECTION-FINDING; RADIO NAVIGATION; DETERMINING DISTANCE OR VELOCITY BY USE OF RADIO WAVES; LOCATING OR PRESENCE-DETECTING BY USE OF THE REFLECTION OR RERADIATION OF RADIO WAVES; ANALOGOUS ARRANGEMENTS USING OTHER WAVES
    • G01S13/00—Systems using the reflection or reradiation of radio waves, e.g. radar systems; Analogous systems using reflection or reradiation of waves whose nature or wavelength is irrelevant or unspecified
    • G01S13/88—Radar or analogous systems specially adapted for specific applications
    • G01S13/93—Radar or analogous systems specially adapted for specific applications for anti-collision purposes
    • G01S13/931—Radar or analogous systems specially adapted for specific applications for anti-collision purposes of land vehicles

Abstract

The distributed radar imaging method based on the MIAA provided by the invention utilizes an MIMO form on the basis of a frequency modulation continuous wave radar and controls the distance between the radars by separately arranging a plurality of radars. Acquiring sampling data of an echo signal of each radar; determining lost data in the same format as the sampled data; carrying out initial spectrum estimation on the sampled data to obtain a frequency spectrum of the sampled data; determining the linear relation between the lost data and the sampled data based on the linear relation between the amplitude spectrum in the frequency spectrum and the sampled data under the condition that the interference covariance matrix of the complete data is known; on the basis of the linear relation between the lost data and the sampled data, the data value in the lost data is estimated by using the minimum mean square error criterion so as to complement the missing part of data, so that the effect of the missing part of data is equal to that of a large-aperture radar, the high-resolution imaging of the radar is further completed, the flexibility of the imaging of the vehicle-mounted radar is improved, and meanwhile, the cost is reduced.

Description

Distributed radar imaging method based on MIAA
Technical Field
The invention belongs to the technical field of radars, and particularly relates to a distributed radar imaging method based on MIAA.
Background
Synthetic Aperture Radar (SAR) can process doppler frequency shift caused by relative motion between a target and the Radar, and effectively improves the resolution of Radar imaging.
The SAR is divided according to waveforms and can be divided into a pulse system SAR and a continuous wave system SAR. The pulse system SAR is a common SAR system, and transmits SAR-related pulse signals at certain time intervals, but the radar of the pulse system has high power consumption and is easy to intercept, so that the application of the pulse system SAR is limited.
High resolution of radar imaging is always a continuous pursuit of many radar researchers, the SAR can create a virtual aperture by establishing relative motion of a target and a radar so as to improve the resolution of the radar, but most of the existing SAR are loaded on a satellite or a plane which is large equipment, the radar system is high in manufacturing cost and long in time consumption, and the situation is very unfavorable for civil use and greatly limits the development of the SAR. And the SAR imaging requires the establishment of relative motion between the target and the radar, which makes SAR imaging have great limitations in some specific high resolution imaging environments, such as vehicle-mounted radar imaging.
Disclosure of Invention
In order to solve the problems in the prior art, the invention provides a distributed radar imaging method based on MIAA. The technical problem to be solved by the invention is realized by the following technical scheme:
the invention provides a distributed radar imaging method based on MIAA, which is applied to a radar system comprising a plurality of radars and comprises the following steps:
acquiring sampling data of an echo signal of each radar;
determining missing data of the same format as the sampled data;
wherein, the data value in the lost data is unknown, and the sampling data and the lost data form the complete data of the echo signal;
performing initial spectrum estimation on the sampled data to obtain a frequency spectrum of the sampled data;
determining the linear relation between the lost data and the sampled data based on the linear relation between the amplitude spectrum in the frequency spectrum and the sampled data under the condition that the interference covariance matrix of the complete data is known;
estimating a data value in the lost data by using a minimum mean square error criterion on the basis of a linear relation between the lost data and the sampled data;
and restoring the estimated lost data and the sampled data into an echo signal.
Optionally, performing initial spectrum estimation on the sample data, and obtaining a spectrum of the sample data includes:
setting Fourier vectors of the sampled data and the lost data at sampling frequency points;
forming the Fourier vectors into a matrix;
representing the echo signal as a first expression comprising the matrix and complete data;
wherein the complete data comprises sampled data and missing data;
calculating an interference covariance matrix of the sampling data at each sampling frequency point;
based on the interference covariance matrix, calculating the spectrum estimation of the sampling data at each sampling frequency point by using a weighted least square criterion;
determining the amplitude of the sampling data at the sampling frequency points based on the spectrum estimation of each sampling frequency point;
and forming a frequency spectrum by the spectrum estimation of each sampling frequency point and the corresponding amplitude.
Wherein the sample data is represented as:
Figure BDA0003102037720000031
the missing data is represented as:
Figure BDA0003102037720000032
the echo signal is represented as:
Figure BDA0003102037720000033
wherein, { tnDenotes the sampling time at which the data is sampled,
Figure BDA0003102037720000034
sample time representing missing dataAnd N represents the length of the sample data,
Figure BDA0003102037720000035
indicating the length of the missing data.
Optionally, the fourier vector of the sampling frequency point of the sampling data is represented as:
Figure BDA0003102037720000036
the fourier vector of the sample bins for missing data is represented as:
Figure BDA0003102037720000037
the fourier vector composition matrix is represented as:
A＝[a(ω1) a(ω2) … a(ωK)]
the first expression is:
y＝Aα(ωk)+ε
wherein, ω isk2 pi K/K denotes the sampling frequency points of the sample data, K1, 2, …, K,
Figure BDA0003102037720000038
vector α ═ α1 α2 … αK]TFor the magnitude of the sample bin, ε represents the noise.
The interference covariance matrix of the sampling data at each sampling frequency point is expressed as:
Figure BDA0003102037720000039
Figure BDA00031020377200000310
the spectral estimation of the sampled data at each sampling frequency point is expressed as:
Figure BDA00031020377200000311
the amplitude of the sampled data at the sampling frequency point is expressed as:
Figure BDA0003102037720000041
wherein R isuDenotes the covariance matrix of the sampled data, and H denotes the conjugate transpose.
Optionally, the estimating, based on the linear relationship between the lost data and the sampled data, a data value in the lost data using a minimum mean square error criterion includes:
determining a linear expression represented by the lost data and the sampled data on the basis of the linear relation between the lost data and the sampled data;
calculating a cross-covariance matrix of the missing data and the sampled data based on the linear expression;
calculating a mean square error of the missing data based on the cross-covariance matrix;
converting the linear coefficient in the linear expression into a vector matrix under the condition of minimum mean square error;
and predicting data values in the lost data based on the vector matrix.
Optionally, the linear expression represented by the missing data estimation value and the sampled data is represented as:
Figure BDA0003102037720000042
the cross-covariance matrix is expressed as:
Figure BDA0003102037720000043
the minimum mean square error of the missing data is expressed as:
Figure BDA0003102037720000044
wherein E {. denotes an expectation operation,
Figure BDA0003102037720000045
tr {. denotes the trace of the matrix, the vector matrix is:
Figure BDA0003102037720000046
data values representing estimates of missing data.
Wherein the data values in the estimated lost data are expressed as:
Figure BDA0003102037720000051
according to the MIAA-based distributed radar imaging method, the MIMO form is utilized, multiple radars are arranged in a split mode, the distance between the radars is controlled, the purpose of controlling the aperture size to acquire data of an ideal target is achieved, missing data are completed by the MIAA method, the effect is equal to that of a large-aperture radar, high-resolution imaging of the radar is further completed, and distributed radar imaging essentially belongs to real-aperture radar imaging. The invention establishes a large-aperture virtual radar for high-resolution imaging, which not only can improve the imaging flexibility of the vehicle-mounted radar, but also can reduce the cost and effectively improve the radar azimuth resolution. Compared with synthetic aperture radar imaging, distributed radar imaging does not need to establish relative motion between a radar and a target, has advantages when used in certain special scenes, can effectively solve the problems that a real aperture radar system is too large and too high in cost, and also effectively solves the problem that the synthetic aperture radar system is limited due to the fact that relative motion between the radar and the target needs to be established.
The present invention will be described in further detail with reference to the accompanying drawings and examples.
Drawings
FIG. 1 is a schematic diagram of a radar imaging process provided by the present invention;
fig. 2 is a schematic flowchart of a MIAA-based distributed radar imaging method according to an embodiment of the present invention;
FIG. 3a shows the distance l between two radars1A distributed imaging result map;
FIG. 3b shows the distance l between two radar beams according to the present invention2A distributed imaging result map;
FIG. 3c shows the distance l between two radar beams according to the present invention3A distributed imaging result map;
FIG. 4a shows the distance l between two radars3A distributed imaging result map;
FIG. 4b shows the distance l between two radar beams according to the present invention4A distributed imaging result map;
FIG. 4c shows the distance l between two radar beams according to the present invention5And (5) a distributed imaging result graph.
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.
Example one
Referring to fig. 1, fig. 1 is a schematic diagram of a radar imaging process provided by the present invention, and the imaging process in fig. 1 mainly includes the steps of:
step 1, performing distance dimension FFT and orientation dimension FFT on original data acquired by the radar.
And 2, selecting the interested target from the angle.
And 3, moving the target of interest selected in the step 2 to 0 degrees through cyclic shift.
And 4, performing low-pass filtering processing to filter the interference on two sides of the target.
And 5, extracting the array element where the target is located.
And 6, supplementing missing data of the two radars by using MIAA.
And 7, performing 2-dimensional FFT on the data to finish imaging.
In the imaging process, the invention provides the MIAA-based distributed radar imaging method to fill up missing data.
As shown in fig. 2, the present invention provides a distributed radar imaging method based on MIAA, which is applied to a radar system including multiple radars, and includes:
s1, acquiring sampling data of the echo signal of each radar;
s2, determining lost data with the same format as the sampling data;
the data value in the lost data is unknown, and the sampled data and the lost data form complete data of an echo signal;
it will be appreciated that missing data is data that is missing when sampling areas not involved by radar in a radar system.
S3, performing initial spectrum estimation on the sampled data to obtain a frequency spectrum of the sampled data;
s4, determining the linear relation between the lost data and the sampled data based on the linear relation between the amplitude spectrum and the sampled data in the frequency spectrum under the condition that the interference covariance matrix of the complete data is known;
s5, on the basis of the linear relation between the lost data and the sampling data, estimating the data value in the lost data by using the minimum mean square error criterion;
and S6, restoring the estimated lost data and the sampled data into an echo signal.
The distributed radar imaging method based on the MIAA provided by the invention utilizes an MIMO form on the basis of a frequency modulation continuous wave radar and controls the distance between the radars by separately arranging a plurality of radars. Acquiring sampling data of an echo signal of each radar; determining lost data in the same format as the sampled data; carrying out initial spectrum estimation on the sampled data to obtain a frequency spectrum of the sampled data; determining the linear relation between the lost data and the sampled data based on the linear relation between the amplitude spectrum in the frequency spectrum and the sampled data under the condition that the interference covariance matrix of the complete data is known; on the basis of the linear relation between the lost data and the sampled data, the data value in the lost data is estimated by using the minimum mean square error criterion so as to complement the missing part of data, so that the effect of the missing part of data is equal to that of a large-aperture radar, the high-resolution imaging of the radar is further completed, the flexibility of the imaging of the vehicle-mounted radar is improved, and meanwhile, the cost is reduced.
As an optional implementation manner of the present invention, performing initial spectrum estimation on the sampled data to obtain a spectrum of the sampled data includes:
the method comprises the following steps: setting Fourier vectors of the sampled data and the lost data at sampling frequency points;
step two: forming a matrix by the Fourier vectors;
step three: representing the echo signal as a first expression comprising a matrix and complete data;
wherein the complete data comprises sampled data and lost data;
step four: calculating an interference covariance matrix of the sampling data at each sampling frequency point;
step five: based on the interference covariance matrix, calculating the spectrum estimation of the sampling data at each sampling frequency point by using a weighted least square criterion;
step six: determining the amplitude of the sampling data at the sampling frequency points based on the spectrum estimation of each sampling frequency point;
step seven: and forming a frequency spectrum by the spectrum estimation of each sampling frequency point and the corresponding amplitude.
Wherein the sample data is represented as:
Figure BDA0003102037720000081
the missing data is represented as:
Figure BDA0003102037720000082
the echo signal is represented as:
Figure BDA0003102037720000083
wherein, { tnDenotes the sampling time at which the data is sampled,
Figure BDA0003102037720000084
indicating the sample time of the missing data, N indicating the number of samplesDepending on the length of the beam, the beam length,
Figure BDA0003102037720000085
indicating the length of the missing data.
Sampling data at a sampling frequency point omegakThe fourier vector of (a) is represented as:
Figure BDA0003102037720000086
the fourier vector of the sample bins for missing data is represented as:
Figure BDA0003102037720000087
the fourier vector composition matrix is represented as:
A＝[a(ω1) a(ω2) … a(ωK)]
the first expression is:
y＝Aα(ωk)+ε
wherein, ω isk2 pi K/K denotes the sampling frequency points of the sample data, K1, 2, …, K,
Figure BDA0003102037720000088
vector α ═ α1 α2 … αK]TEpsilon represents a constant for the magnitude of the sampled bins.
For sampling data, each frequency point omegakInterference covariance matrix Q ofu(ωk) Is defined as:
Figure BDA0003102037720000091
Figure BDA0003102037720000092
using a weighted least squares criterion, the sampled data is at each sampleFrequency point omegakThe spectral estimate of (a) is expressed as:
Figure BDA0003102037720000093
from which the amplitude can be estimated
Figure BDA0003102037720000094
Is composed of
Figure BDA0003102037720000095
The above equation is simplified, and the amplitude of the sampling data at the sampling frequency point is represented as:
Figure BDA0003102037720000096
wherein R isuDenotes the covariance matrix of the sampled data, and H denotes the conjugate transpose.
As an alternative embodiment of the present invention, estimating the data value in the missing data by using the minimum mean square error criterion based on the linear relationship between the missing data and the sampled data includes:
the method comprises the following steps: determining a linear expression represented by the lost data and the sampled data on the basis of the linear relation between the lost data and the sampled data;
step two: calculating a cross covariance matrix of the lost data and the sampled data based on the linear expression;
step three: calculating the mean square error of the lost data based on the cross covariance matrix;
step four: converting the linear coefficient in the linear expression into a vector matrix under the condition of minimum mean square error;
step five: data values in the missing data are predicted based on the vector matrix.
The amplitude of the sampling data at the sampling frequency point indicates that when the matrix RuGiven the amplitude estimated by IAADegree spectrum
Figure BDA0003102037720000097
And yuIs linear. Thus, assume matrix RuAnd the covariance matrix R of the entire data is given, yuAnd ymAlso in a linear function relationship. Because of yuAnd ymThere is a linear relationship between them, so a linear Minimum Mean Square Error criterion (MMSE) can be used, by using yuTo estimate ym。
Definition T is
Figure BDA0003102037720000101
The linear expression of the missing data estimate and the sample data representation is:
Figure BDA0003102037720000102
the cross-covariance matrix is expressed as:
Figure BDA0003102037720000103
the minimum mean square error of the missing data is expressed as:
Figure BDA0003102037720000104
wherein E {. denotes an expectation operation,
Figure BDA0003102037720000105
tr {. denotes the trace of the matrix, the vector matrix is:
Figure BDA0003102037720000106
Figure BDA0003102037720000107
data values representing estimates of missing data。
Wherein, the data value in the estimated lost data is represented as:
Figure BDA0003102037720000108
the effect of the imaging method provided by the invention is further verified by combining simulation experiments.
1. And (5) simulating experimental conditions.
The hardware platform of the simulation experiment of the invention is as follows: the TI official cascade imaging radar system consists of two parts, wherein one part is a radio frequency board MMWCAS-RF-EVM, and the other part is a data acquisition storage board MMWCAS-DSP-EVM.
The radio frequency development board is provided with an AWR2243 radio frequency chip, and a phase shifter, an LNA, a baseband signal conditioning circuit, an AD acquisition circuit and the like are integrated in the chip. The AWR2243 radio frequency chip works in the frequency range of 76-81GHz, the maximum bandwidth is 5GHz, and the output power of TX is 13 dBm.
The AWR2243 chip contains a radio frequency front end control circuit and an AD acquisition circuit, and a radio frequency front end with three transmitting chains and four receiving chains can form a 3-transmitting 4-receiving MIMO array, wherein each transmitter includes a programmable 6-bit phase shifter (step by 5.625 °) for beam forming application. Four receiving link signals and transmitting signals in an AWR2243 radar chip are subjected to frequency mixing through a frequency mixer to obtain intermediate frequency signals, then after a series of signal conditioning such as signal amplification and filtering, IQ sampling is carried out on four paths of echo signals through an ADC, and parameters such as sampling frequency and sampling point number can be configured through mmwave studio software. The sampled data is stored in the ADCbuffer, and then the data is taken out from the ADCbuffer for subsequent processing
4 AWR2243 radar chips are cascaded on the MMWCAS-RF-EVM, one of the chips is a master chip, the other three chips are slave chips, the radio frequency board is provided with 12 transmitting antennas and 16 receiving antennas in a cascading mode to form a 12-transmitting and 16-receiving MIMO array, and the resolution ratio of the radar can be effectively improved. The angular resolution of a typical front-mounted radar sensor is about 5 °, and the radar sensors after the cascade can achieve an angular resolution of 1.4 °.
The cascaded radio frequency boards MMWCAS-RF-EVM are provided with an antenna array formed by a master device and three slave devices. Of the 12 transmit antennas, 9 are placed in parallel for azimuth estimation, 3 are placed vertically for elevation estimation, and all 16 receive antennas are placed horizontally. Every two transmitting antennas are spaced by 2 wavelengths, and every two receiving antennas and all formed virtual channels are spaced by half wavelengths. RX array a and RX array C are spaced 16 wavelengths apart and RX array C and RX array B are spaced 4 wavelengths apart. In MIMO mode, 86 virtual channels may be formed for position estimation.
The MMWCAS-DSP-EVM of the data acquisition board of the cascade radar system mainly integrates a TDA2x high-performance multimedia application processor, 4 pieces of FPGA, 2Gb DDR3L, a 512Gb SSD and a plurality of interfaces, wherein the TDA2x is used for data preprocessing, and the SSD is used for data storage.
The method comprises the steps of connecting a radio frequency board MMWCAS-RF-EVM and a data storage version MMWCAS-DSP-EVM through an integrated interface to form a complete radar system, wherein the radio frequency board is used for transmitting and receiving signals, the signals pass through a mixer to obtain intermediate frequency signals, then the intermediate frequency signals are sampled through an ADC (analog to digital converter), data preprocessing is carried out through the data storage board and stored, and finally the data is exported to a PC (personal computer) through software winSCP to be used for subsequent research.
The cascaded radar has two modes which can be selected for use, one is a MIMO mode, and the other is a beam forming mode.
In the MIMO mode, each TX channel transmits independently to formulate a large virtual array, thereby achieving high angular resolution. The maximum achievable virtual array in MIMO mode is 86 channels. Orthogonality between all TX channels may be achieved by using time-duplex multiplexing or orthogonal binary sequences. In the research of the invention, the MIMO form is taken as the main mode, only one transmitting antenna is used for transmitting the chirp pulse at a time in a time division multiplexing mode, and all 12 transmitting antennas transmit the chirp signal as a cycle, thereby improving the azimuth resolution.
In TX beamforming mode, multiple TX channels transmit simultaneously and coherently to achieve higher gain and longer detection range in the main focused field of view. Each TX channel of each device has a 6-bit configurable phase register with a step size of 5.625 °. The phase value programmed into each TX channel is calculated from the position where the main beam should be focused. Since only 9 TX channels are located on the same azimuth plane on the TI4 chip-cascade board, these 9 TX channels can be used to steer the beam. Because coherent gain is achieved in the TX beamforming mode, its detection range is much longer than that achievable in the MIMO mode.
The software platform of the simulation experiment of the invention is as follows: mmwave studio and MATLAB R2018 b.
2. And (5) analyzing simulation contents and results thereof.
The simulation experiment of the invention adopts distributed radar imaging, two radars are respectively arranged on a slide rail according to a certain scale to enlarge the radar aperture and collect data, then missing data is supplemented by using an MIAA algorithm, and finally high-resolution imaging is completed.
The cascaded radar radio frequency board card is cascaded with 4 AWR2243 radar chips, the single AWR2243 chip is provided with 3 transmitting antennas and 4 receiving antennas, so that the cascaded radar is provided with 12 transmitting antennas and 16 receiving antennas, 192 virtual channels can be formed, 9 antennas in the transmitting antennas are used for direction estimation, and due to the fact that the channels are overlapped, 86 virtual channels are used for direction estimation. The channel-to-channel spacing is half a wavelength, since the cascaded radar can achieve 76-81GHz frequency variation, the half wavelength length is 76.78 mils as known by the TI official document. The total length of the MMWCAS-RF-EVM radio frequency board card is 160mm, 86 virtual channels are arranged from the left side of the board card by 24.7mm, and the layout is finished by extending the right side of the board card by 32.3 mm. Therefore, for a single radar, the physical center of the antenna is 106.4mm from the left of the board card.
According to the size of the MMWCAS-RF-EVM and the actual arrangement of the virtual arrays, 14 array elements can be arranged from the left edge of the board card to the start of the arrays, and 15 array elements can be arranged from the end of the array arrangement to the right edge of the board card. If will get up the virtual channel seamless connection of two radars, because the reason of integrated circuit board size, need reserve 4 half wavelength's distance between two radars, because every radar has a metal crate to be used for setting up the radar again, when two metal crates are close-fitting, two radars interval 11.5mm need take this part into account when calculating distributed radar aperture size and separately putting the radar on the slide rail.
When the aperture is expanded, in order to enable the two real radar virtual arrays and the array of the vacant part to be seamlessly connected, the positions of the radars need to be strictly adjusted, and the distance between the two radars is calculated as the following formula, wherein the distance is represented as lnThen there is
ln＝(86*(n-1)+4)*λ/2
Where n is 1, …, and 5, the number of radar arrays from the absence to the absence 4 is sequentially indicated, and λ/2 is a half wavelength. When n is 1, the distance between the two radar frames is 4 half wavelengths, but because the distance between the two radars is more than 4 half wavelengths when the radar metal frames are tightly attached, the distance between the two radars when the radar arrays are not lacked between the two radars, namely the distance between the two radars when the two radar metal frames are tightly attached, namely 11.5 mm.
TABLE 1 different Aperture distributed radars in slide position
Figure BDA0003102037720000141
Table 1 shows the situation where different aperture distributed radars are placed at different positions of the slide rail. In the research of distributed radar imaging, two cascaded imaging radars need to be used at the same time, at the moment, the two radars need to be distinguished, the distinguishing is to adopt a frequency division mode to enable the two radars to use different frequency bands in the working process, and a certain protection bandwidth is reserved between the frequency bands used by the two radars. The parameter settings of the two radars in the inventive experiment are shown in tables 2, 3, 4 and 5. Tables 2 and 3 show the parameter settings for the radar operating in the dark room, and tables 4 and 5 show the parameter settings for the radar operating outside.
TABLE 2 frequency division mode parameter settings (left radar)
Parameter(s) Starting frequency Slope of frequency modulation Sampling rate Bandwidth of Pulse width Sampling point
Parameter value 76GHz 40MHz/us 5M 1G 25us 100
TABLE 3 frequency division mode parameter settings (Right Radar)
Parameter(s) Starting frequency Slope of frequency modulation Sampling rate Bandwidth of Pulse width Sampling point
Parameter value 77.5GHz 40MHz/us 5M 1G 25us 100
TABLE 4 frequency division mode parameter settings (left Radar)
Parameter(s) Starting frequency Slope of frequency modulation Sampling rate Bandwidth of Pulse width Sampling point
Parameter value 76GHz 30MHz/us 8M 1.5G 50us 300
TABLE 5 Subdivision mode parameter settings (Right Radar)
Parameter(s) Starting frequency Slope of frequency modulation Sampling rate Bandwidth of Pulse width Sampling point
Parameter value 78GHz 30MHz/us 8M 1.5G 50us 300
3. Analysis of simulation results
The effects of the present invention will be further described with reference to the experimental result chart.
FIG. 3a shows the distance l between two radars1And (5) distributed imaging results. FIG. 3b shows the distance l between two radars2And (5) distributed imaging results. FIG. 3c shows the distance l between two radars3And (5) distributed imaging results.
FIG. 4a shows the distance l between two radars3And (5) distributed imaging results. FIG. 4b shows the distance l between two radars4And (5) distributed imaging results. FIG. 4c shows the distance l between two radars5And (5) distributed imaging results.
The simulation results analysis shown in fig. 3a to 3c described above are results analysis with the abscissa as azimuth/° and the ordinate as amplitude.
The simulation results analysis shown in fig. 4a to 4c described above are for results analysis with the abscissa as azimuth/° and the ordinate as distance/m.
As can be seen from fig. 3a to 3c, the azimuth resolution of the radar gradually increases with the increase of the distributed radar aperture in the frequency division mode. As can be seen from fig. 4a to 4c, in the frequency division mode, the radar has better and better imaging effect on the object as the aperture of the distributed radar increases.
The above experiments show that: the MIAA-based distributed radar imaging method can clearly image a target, effectively improves the radar azimuth resolution and the imaging effect, and verifies the feasibility and the effectiveness of distributed radar imaging through the imaging result of an actual measurement scene.
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.

Claims (8)

1. A distributed radar imaging method based on MIAA is applied to a radar system comprising a plurality of radars, and is characterized by comprising the following steps:
acquiring sampling data of an echo signal of each radar;
determining missing data of the same format as the sampled data;
wherein, the data value in the lost data is unknown, and the sampling data and the lost data form the complete data of the echo signal;
performing initial spectrum estimation on the sampled data to obtain a frequency spectrum of the sampled data;
determining the linear relation between the lost data and the sampled data based on the linear relation between the amplitude spectrum in the frequency spectrum and the sampled data under the condition that the interference covariance matrix of the complete data is known;
estimating a data value in the lost data by using a minimum mean square error criterion on the basis of a linear relation between the lost data and the sampled data;
and restoring the estimated lost data and the sampled data into an echo signal.
2. The method of claim 1, wherein performing an initial spectrum estimation on the sampled data to obtain a spectrum of the sampled data comprises:
setting Fourier vectors of the sampled data and the lost data at sampling frequency points;
forming the Fourier vectors into a matrix;
representing the echo signal as a first expression comprising the matrix and complete data;
wherein the complete data comprises sampled data and missing data;
calculating an interference covariance matrix of the sampling data at each sampling frequency point;
based on the interference covariance matrix, calculating the spectrum estimation of the sampling data at each sampling frequency point by using a weighted least square criterion;
determining the amplitude of the sampling data at the sampling frequency points based on the spectrum estimation of each sampling frequency point;
and forming a frequency spectrum by the spectrum estimation of each sampling frequency point and the corresponding amplitude.
3. The method of claim 2, wherein the sample data is represented as:
Figure FDA0003102037710000021
the missing data is represented as:
Figure FDA0003102037710000022
the echo signal is represented as:
Figure FDA0003102037710000023
wherein, { tnDenotes the sampling time at which the data is sampled,
Figure FDA0003102037710000024
indicating the sample time of the missing data, N indicating the sample data length,
Figure FDA0003102037710000025
indicating the length of the missing data.
4. The method of claim 3, wherein the Fourier vector of the sampling frequency points of the sampled data is represented as:
Figure FDA0003102037710000026
the fourier vector of the sample bins for missing data is represented as:
Figure FDA0003102037710000027
the fourier vector composition matrix is represented as:
A＝[a(ω1) a(ω2)…a(ωK)]
the first expression is:
y＝Aα(ωk)+ε
wherein, ω isk2 pi K/K denotes the sampling frequency points of the sample data, K1, 2, …, K,
Figure FDA0003102037710000028
vector α ═ α1 α2…αK]TFor the magnitude of the sample bin, ε represents the noise.
5. The method of claim 4, wherein the interference covariance matrix of the sampled data at each sampling bin is represented as:
Figure FDA0003102037710000029
Figure FDA00031020377100000210
the spectral estimation of the sampled data at each sampling frequency point is expressed as:
Figure FDA0003102037710000031
the amplitude of the sampled data at the sampling frequency point is expressed as:
Figure FDA0003102037710000032
wherein R isuDenotes the covariance matrix of the sampled data, and H denotes the conjugate transpose.
6. The method of claim 5, wherein estimating the data values in the missing data using a minimum mean square error criterion based on a linear relationship of the missing data to the sampled data comprises:
determining a linear expression represented by the lost data and the sampled data on the basis of the linear relation between the lost data and the sampled data;
calculating a cross-covariance matrix of the missing data and the sampled data based on the linear expression;
calculating a mean square error of the missing data based on the cross-covariance matrix;
converting the linear coefficient in the linear expression into a vector matrix under the condition of minimum mean square error;
and predicting data values in the lost data based on the vector matrix.
7. The method of claim 6, wherein the linear representation of the missing data estimate and the sampled data representation is represented as:
Figure FDA0003102037710000033
the cross-covariance matrix is expressed as:
Figure FDA0003102037710000034
the minimum mean square error of the missing data is expressed as:
Figure FDA0003102037710000041
wherein E {. denotes an expectation operation,
Figure FDA0003102037710000042
tr {. denotes the trace of the matrix, the vector matrix is:
Figure FDA0003102037710000043
Figure FDA0003102037710000044
data values representing estimates of missing data.
8. The method of claim 7, wherein the data values in the predicted loss data are expressed as:
Figure FDA0003102037710000045
CN202110627285.1A 2021-06-04 2021-06-04 Distributed radar imaging method based on MIAA Pending CN113534149A (en)

Priority Applications (1)

Application Number Priority Date Filing Date Title
CN202110627285.1A CN113534149A (en) 2021-06-04 2021-06-04 Distributed radar imaging method based on MIAA

Applications Claiming Priority (1)

Application Number Priority Date Filing Date Title
CN202110627285.1A CN113534149A (en) 2021-06-04 2021-06-04 Distributed radar imaging method based on MIAA

Publications (1)

Publication Number Publication Date
CN113534149A true CN113534149A (en) 2021-10-22

Family

ID=78095625

Family Applications (1)

Application Number Title Priority Date Filing Date
CN202110627285.1A Pending CN113534149A (en) 2021-06-04 2021-06-04 Distributed radar imaging method based on MIAA

Country Status (1)

Country Link
CN (1) CN113534149A (en)

Citations (6)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN103983959A (en) * 2014-05-16 2014-08-13 西安电子科技大学 SAR system movement target radial speed estimation method based on data reconstruction
CN104020469A (en) * 2014-05-30 2014-09-03 哈尔滨工程大学 MIMO radar distance-angle two-dimensional super-resolution imaging algorithm
CN104991237A (en) * 2015-06-25 2015-10-21 西安电子科技大学 Steady space-time adaptive processing-based target detection method
CN105785361A (en) * 2016-03-08 2016-07-20 南京信息工程大学 MIMO radar imaging method on condition of array element failure
CN106199602A (en) * 2016-08-31 2016-12-07 电子科技大学 A kind of Staggered SAR echo signal method for reconstructing
CN108318879A (en) * 2017-12-29 2018-07-24 哈尔滨工业大学 ISAR image transverse direction calibrating methods based on IAA Power estimation technologies

Patent Citations (6)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN103983959A (en) * 2014-05-16 2014-08-13 西安电子科技大学 SAR system movement target radial speed estimation method based on data reconstruction
CN104020469A (en) * 2014-05-30 2014-09-03 哈尔滨工程大学 MIMO radar distance-angle two-dimensional super-resolution imaging algorithm
CN104991237A (en) * 2015-06-25 2015-10-21 西安电子科技大学 Steady space-time adaptive processing-based target detection method
CN105785361A (en) * 2016-03-08 2016-07-20 南京信息工程大学 MIMO radar imaging method on condition of array element failure
CN106199602A (en) * 2016-08-31 2016-12-07 电子科技大学 A kind of Staggered SAR echo signal method for reconstructing
CN108318879A (en) * 2017-12-29 2018-07-24 哈尔滨工业大学 ISAR image transverse direction calibrating methods based on IAA Power estimation technologies

Non-Patent Citations (2)

* Cited by examiner, † Cited by third party
Title
PETRE STOICA ET AL.: "Missing Data Recovery Via a Nonparametric Iterative Adaptive Approach", 《IEEE SIGNAL PROCESSING LETTERS》 *
邹江波: "捷变PRF合成孔径雷达数据重建方法研究", 《中国优秀博硕士学位论文全文数据库(硕士) 信息科技辑》 *

Similar Documents

Publication Publication Date Title
EP3471210B1 (en) Radar apparatus
CN112578340B (en) Radar device and radar signal processing method
Brenner et al. Demonstration of advanced reconnaissance techniques with the airborne SAR/GMTI sensor PAMIR
EP2798370B1 (en) Coded aperture beam analysis method and apparatus
US20170276769A1 (en) Radar apparatus and radar method
DE102019125973A1 (en) Radar device
WO2015126505A2 (en) Methods and apparatus for reducing noise in a coded aperture radar
CN108387877B (en) Moving target phase correction method of multi-input multi-output radar
WO2013141924A2 (en) Coded aperture beam analysis method and apparatus
CN110412570B (en) HRWS-SAR imaging method based on spatial pulse phase coding
CN111090080B (en) Ultra-wideband radar single-channel digital beam forming method based on space-time coding array
DE102019125502A1 (en) Radar device and targeting method
CN104777467B (en) Object detection method based on frequency scan antenna
CN108693526B (en) Moving object phase compensation method for MIMO radar
DE112020001327T5 (en) Radar device
DE112020001356T5 (en) Radar device and transmitter / receiver group antenna
CN113534149A (en) Distributed radar imaging method based on MIAA
Kim et al. Investigation of a new multifunctional high performance SAR system concept exploiting MIMO technology
Wang et al. Improving the image quality of spaceborne multiple-aperture SAR under minimization of sidelobe clutter and noise
Lynch Low latency digital beamforming radar using aperture coding
CN110658502A (en) Amplitude-phase error correction method
CN111443336B (en) Method for reducing data transmission throughput of FMCW radar system
Anajemba et al. Efficient switched digital beamforming radar system based on SIMO/MIMO receiver
CN114185018A (en) Phase noise digital real-time estimation method and system of frequency modulation continuous wave radar
Takayama et al. Hybrid SIMO and MIMO sparse array radar

Legal Events

Date Code Title Description
PB01 Publication
PB01 Publication
SE01 Entry into force of request for substantive examination
SE01 Entry into force of request for substantive examination
RJ01 Rejection of invention patent application after publication
RJ01 Rejection of invention patent application after publication

Application publication date: 20211022