CN116184332B - Radar interference suppression method, device and storage medium - Google Patents

Radar interference suppression method, device and storage medium Download PDF

Info

Publication number
CN116184332B
CN116184332B CN202310433569.6A CN202310433569A CN116184332B CN 116184332 B CN116184332 B CN 116184332B CN 202310433569 A CN202310433569 A CN 202310433569A CN 116184332 B CN116184332 B CN 116184332B
Authority
CN
China
Prior art keywords
dimensional
signal
radar
function
interference
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.)
Active
Application number
CN202310433569.6A
Other languages
Chinese (zh)
Other versions
CN116184332A (en
Inventor
黄岩
王韵旋
毛源
刘江
杨阳
张慧
洪伟
冯友怀
郭坤鹏
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.)
Southeast University
Nanjing Hawkeye Electronic Technology Co Ltd
Original Assignee
Southeast University
Nanjing Hawkeye Electronic Technology Co Ltd
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 Southeast University, Nanjing Hawkeye Electronic Technology Co Ltd filed Critical Southeast University
Priority to CN202310433569.6A priority Critical patent/CN116184332B/en
Publication of CN116184332A publication Critical patent/CN116184332A/en
Application granted granted Critical
Publication of CN116184332B publication Critical patent/CN116184332B/en
Active legal-status Critical Current
Anticipated expiration legal-status Critical

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
    • G01S7/00—Details of systems according to groups G01S13/00, G01S15/00, G01S17/00
    • G01S7/02—Details of systems according to groups G01S13/00, G01S15/00, G01S17/00 of systems according to group G01S13/00
    • G01S7/36—Means for anti-jamming, e.g. ECCM, i.e. electronic counter-counter measures
    • 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
    • G01S7/00—Details of systems according to groups G01S13/00, G01S15/00, G01S17/00
    • G01S7/02—Details of systems according to groups G01S13/00, G01S15/00, G01S17/00 of systems according to group G01S13/00
    • G01S7/41—Details of systems according to groups G01S13/00, G01S15/00, G01S17/00 of systems according to group G01S13/00 using analysis of echo signal for target characterisation; Target signature; Target cross-section

Landscapes

  • Engineering & Computer Science (AREA)
  • Computer Networks & Wireless Communication (AREA)
  • Physics & Mathematics (AREA)
  • General Physics & Mathematics (AREA)
  • Radar, Positioning & Navigation (AREA)
  • Remote Sensing (AREA)
  • Radar Systems Or Details Thereof (AREA)

Abstract

The invention discloses a radar interference suppression methodThe method, the device and the storage medium, the method comprises the following steps: constructing a three-dimensional radar signal with a slow time dimension, a fast time dimension and a channel dimension according to the radar echo signal; constraining a target signal in a three-dimensional radar signal with a tensor kernel norm, using a tensor
Figure ZY_1
Norm constrains an interfering signal in the three-dimensional radar signal; based on tensor kernel norms and tensors
Figure ZY_2
Establishing a target function by using norms, and establishing a constraint function based on the three-dimensional radar signals, the target signals and the interference signals so as to establish an interference suppression model; and solving the interference suppression model by an alternate direction multiplier method to obtain a real target signal after interference suppression. The technical scheme provided by the invention can solve the technical problem of low evaluation accuracy of the target signal when the radar interference signal is detected in the prior art, and is suitable for radar signals with nonstandard interference energy and longer interference duration.

Description

Radar interference suppression method, device and storage medium
Technical Field
The present invention relates to the field of radar technologies, and in particular, to a method and apparatus for suppressing radar interference, and a storage medium.
Background
The vehicle millimeter wave radar (automotive millimeter wave radar) is a sensor dedicated to sensing an external environment, assisting an advanced driving assistance system, and generating necessary driving instructions in real time according to a road environment. Compared with other similar sensors such as cameras and laser radars, the vehicle millimeter wave radar has the working characteristics of all-day and all-weather, can obtain useful information such as the distance, the speed, the angle and the like of a target, has low manufacturing cost, and is an important civil radar system. Meanwhile, because of the development of the chip industry, the cost of a Multiple Input Multiple Output (MIMO) radar system is remarkably reduced, and because of the capability of angle estimation, the method has great significance for resolving speed ambiguity. However, since the bandwidth used by the vehicle-mounted millimeter wave radar is limited to 77-81GHz, mutual interference between the vehicle-mounted millimeter wave radars is unavoidable. Because the interference signal directly reaches the radar receiving antenna, the middle part is not reflected by a target, the energy is usually 20-30dB higher than that of the useful signal, the interference signal can obviously reduce the signal-to-interference-noise ratio of the echo, and the whole image is caused to present suppressed pollution due to the lack of the useful echo signal, so that the RD image texture is completely invisible.
The existing interference detection method mainly depends on energy change, and the energy of radio frequency interference is 20dB to 40dB higher than that of useful signals, so that the method for detecting the existence of interference by using an energy threshold detection method is not suitable for special situations such as nonstandard interference energy or longer interference duration, and is especially not suitable for application scenes with higher requirements on target estimation accuracy.
Disclosure of Invention
The invention provides a radar interference suppression method, a radar interference suppression device and a storage medium, which aim to effectively solve the technical problem that in the prior art, when radar interference signals are detected, target signal evaluation accuracy is low.
According to an aspect of the present invention, there is provided a method for suppressing radar interference, for a MIMO radar including a plurality of transmitting antennas and a plurality of receiving antennas, the plurality of transmitting antennas and the plurality of receiving antennas constituting a plurality of channels, the method comprising:
acquiring radar echo signals obtained by the radar in the multiple channels, and constructing three-dimensional radar signals in a slow time dimension, a fast time dimension and a channel dimension according to the radar echo signals;
constraining a target signal in the three-dimensional radar signal with a tensor kernel norm, using a tensor
Figure SMS_1
Norms constrain interfering signals in the three-dimensional radar signal;
based on the tensor kernel norms and the tensor
Figure SMS_2
Establishing an objective function by using norms, establishing a constraint function based on the three-dimensional radar signals, the objective signals and the interference signals, and establishing an interference suppression model according to the objective function and the constraint function;
and solving the interference suppression model by an alternate direction multiplier method to obtain a real target signal after interference suppression.
Further, the tensor-based kernel norms and the tensors
Figure SMS_3
The norm construction objective function includes:
the objective function is constructed according to the following equation:
Figure SMS_4
，
wherein,,
Figure SMS_5
representing the target signal,/->
Figure SMS_6
Representing the interference signal,/->
Figure SMS_7
Representing a first superparameter,/->
Figure SMS_8
Representing tensor kernel norms,/->
Figure SMS_9
Representing tensor->
Figure SMS_10
Norms (F/F)>
Figure SMS_11
Indicating that the parameters in brackets are minimized.
Further, the constructing a constraint function based on the three-dimensional radar signal, the target signal, and the interference signal includes:
constructing the constraint function according to the following formula:
Figure SMS_12
，
wherein,,
Figure SMS_13
representing the three-dimensional radar signal,/->
Figure SMS_14
Representing the target signal,/->
Figure SMS_15
Representing the interference signal,/->
Figure SMS_16
Represents noise margin, ++>
Figure SMS_17
The Frobenius norm of the tensor is represented.
Further, the solving the interference suppression model by the alternate direction multiplier method includes:
constructing an augmented Lagrangian function according to the objective function and the constraint function;
and constructing a threshold function and a soft threshold function, and solving the augmented Lagrangian function by an alternate direction multiplier method.
Further, the constructing an augmented lagrangian function from the objective function and the constraint function includes:
and constructing the augmented Lagrange function according to the objective function and the constraint function by taking the objective signal, the interference signal, the Lagrange variable and the second super parameter as variables.
Further, the constructing a threshold function and a soft threshold function, and solving the augmented lagrangian function by an alternating direction multiplier method includes:
decomposing the three-dimensional radar signals into two-dimensional radar signals corresponding to the channels;
performing multiple iterations on the two-dimensional radar signals of each channel, and calculating an updated two-dimensional target signal closed solution according to a singular value decomposition method and a threshold function in each iteration;
performing inverse Fourier transform on a plurality of two-dimensional target signal closed solutions corresponding to the channels to obtain a target signal three-dimensional tensor;
calculating a two-dimensional interference signal closed solution according to the soft threshold function and each two-dimensional target signal closed solution;
obtaining a three-dimensional tensor of the interference signal according to closed solution of a plurality of two-dimensional interference signals corresponding to the channels;
obtaining a two-dimensional Lagrangian variable closed solution according to the two-dimensional target signal closed solution and the two-dimensional interference signal closed solution;
obtaining a Lagrange variable three-dimensional tensor according to a plurality of two-dimensional Lagrange variable closed solutions corresponding to the channels;
calculating a super-parameter closed solution and calculating a termination condition function;
and stopping iteration when the iteration number reaches the preset iteration number or the value of the termination condition function is smaller than the preset error tolerance.
Further, the performing multiple iterations on the two-dimensional radar signal of each channel, and calculating an updated two-dimensional target signal closed-form solution according to the singular value decomposition method and the threshold function in each iteration includes:
performing Fourier transformation on a target signal in the three-dimensional radar signal to obtain a preprocessed target signal, and obtaining a two-dimensional target signal of each channel according to the preprocessed target signal;
based on the two-dimensional radar signals of each channel and the two-dimensional Lagrange variable closed solution and the two-dimensional interference signal closed Jie Shengcheng first matrix obtained by the last iteration of the channel;
singular value decomposition is carried out on the first matrix to obtain a left singular matrix, a two-dimensional target signal closed solution and a right singular matrix;
and generating an updated two-dimensional target signal closed solution based on the left singular matrix, the two-dimensional target signal closed solution, the right singular matrix and the threshold function.
Further, the calculating the termination condition function includes:
and calculating the termination condition function based on the three-dimensional radar signal, the target signal three-dimensional tensor obtained after iteration, the interference signal three-dimensional tensor, the Lagrangian variable three-dimensional tensor and the super-parameter closed solution.
According to another aspect of the present invention, there is also provided a radar interference suppression apparatus for a MIMO radar including a plurality of transmitting antennas and a plurality of receiving antennas, the plurality of transmitting antennas and the plurality of receiving antennas constituting a plurality of channels, the apparatus comprising:
the signal acquisition module is used for acquiring radar echo signals obtained by the radar in the multiple channels and constructing three-dimensional radar signals in a slow time dimension, a fast time dimension and a channel dimension according to the radar echo signals;
a norm constraint module for constraining the three-dimensional radar signal with tensor kernel normsTarget signal, using tensor
Figure SMS_18
Norms constrain interfering signals in the three-dimensional radar signal;
a model construction module for constructing a model based on the tensor kernel norms and the tensors
Figure SMS_19
Establishing an objective function by using norms, establishing a constraint function based on the three-dimensional radar signals, the objective signals and the interference signals, and establishing an interference suppression model according to the objective function and the constraint function;
and the calculation module is used for solving the interference suppression model through an alternate direction multiplier method so as to obtain a real target signal after interference suppression.
According to another aspect of the present invention there is also provided a storage medium having stored therein a plurality of instructions adapted to be loaded by a processor to perform any of the radar disturbance suppression methods as described above.
Through one or more of the above embodiments of the present invention, at least the following technical effects can be achieved:
in the technical scheme disclosed by the invention, the special edge channel low-rank property of the useful target signal on the three-dimensional time domain matrix and the sparse property of the interference signal are analyzed, and the interference suppression problem is constructed to separate the interference signal with sparse property and the target signal with low-rank property. And solving an interference suppression model through an Alternating Direction Multiplier Method (ADMM) to obtain an interference suppression result. The interference suppression method is suitable for various application scenes with longer interference duration and more interference numbers, and compared with the traditional method, the accuracy of suppressing the interference signals can be obviously improved.
Drawings
The technical solution and other advantageous effects of the present invention will be made apparent by the following detailed description of the specific embodiments of the present invention with reference to the accompanying drawings.
Fig. 1 is a step flowchart of a radar interference suppression method provided in an embodiment of the present invention;
FIG. 2 is a schematic diagram of signals having different characteristics in the three-dimensional time-frequency domain;
FIG. 3 is a range-Doppler plot of a real target signal after interference suppression;
fig. 4 is a schematic structural diagram of a radar interference suppression device according to an embodiment of the present invention.
Detailed Description
The technical solutions in the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings in the embodiments of the present invention. It will be apparent that the described embodiments are only some, but not all, embodiments of the invention. All other embodiments, which can be made by those skilled in the art based on the embodiments of the invention without making any inventive effort, are intended to fall within the scope of the invention.
In the description of the present invention, it should be noted that, unless explicitly specified and defined otherwise, the term "and/or" herein is merely an association relationship describing associated objects, and means that three relationships may exist, for example, a and/or B may mean: a exists alone, A and B exist together, and B exists alone. The character "/" herein generally indicates that the associated object is an "or" relationship unless otherwise specified.
Fig. 1 is a flowchart illustrating steps of a method for suppressing radar interference according to an embodiment of the present invention, where according to an aspect of the present invention, the method for suppressing radar interference is provided, and the method for suppressing radar interference is used for a MIMO radar, where the MIMO radar includes a plurality of transmitting antennas and a plurality of receiving antennas, and the plurality of transmitting antennas and the plurality of receiving antennas form a plurality of channels, and the method for suppressing radar interference includes:
step 101: acquiring radar echo signals obtained by the radar in the multiple channels, and constructing three-dimensional radar signals in a slow time dimension, a fast time dimension and a channel dimension according to the radar echo signals;
step 102: constraining a target signal in the three-dimensional radar signal with a tensor kernel norm, using a tensorMeasuring amount
Figure SMS_20
Norms constrain interfering signals in the three-dimensional radar signal;
step 103: based on the tensor kernel norms and the tensor
Figure SMS_21
Establishing an objective function by using norms, establishing a constraint function based on the three-dimensional radar signals, the objective signals and the interference signals, and establishing an interference suppression model according to the objective function and the constraint function;
step 104: and solving the interference suppression model by an alternate direction multiplier method to obtain a real target signal after interference suppression.
In order to reject the interference and preserve more useful signals, it is necessary to find a specific basis to separate the useful signal from the interference based on the characteristics of the interference and the useful signal. The invention provides a TRPC-based interference suppression method for mutual interference among MIMO vehicle-mounted millimeter wave radars, which is based on TRPC and is provided for mutual interference among the MIMO vehicle-mounted millimeter wave radars using Frequency modulation continuous wave (Frequency-Modulated Continuous Wave, FMCW for short).
And analyzing the low-rank property of the special edge channel of the useful target signal on the three-dimensional time domain matrix and the sparse property of the interference signal on the three-dimensional time domain matrix, and constructing an interference suppression problem to separate the interference signal with sparse characteristic and the useful target signal with low-rank characteristic. The optimization problem is solved by an alternate direction multiplier method (Alternating Direction Method of Multipliers, abbreviated as ADMM) to obtain interference suppression results.
The steps 101 to 104 are specifically described below.
In step 101, radar echo signals obtained by the radar in the multiple channels are obtained, and three-dimensional radar signals in a slow time dimension, a fast time dimension and a channel dimension are constructed according to the radar echo signals.
Illustratively, in addition to conventionally processing the radar echo signals into time domain signals, the channel dimensions are increased, resulting in three-dimensional radar signals. For example, a radar with 3 transmit antennas and 4 receive antennas, with 12 channels formed between the transmit and receive antennas.
In step 102, a target signal in the three-dimensional radar signal is constrained with a tensor kernel norm, with a tensor
Figure SMS_22
The norms constrain interfering signals in the three-dimensional radar signal.
Illustratively, fig. 2 is a schematic diagram of signals having different characteristics in the three-dimensional time-frequency domain, the three-dimensional radar signals representing three-dimensional images of radar echo signals, including interfering signals and useful signals. In order to suppress the interference signal and reserve more target numbers, a specific base is searched for to separate the useful signal from the interference according to different characteristics of the interference signal and the target signal, and a corresponding norm is constructed. The target signal has special channel low-rank property on the three-dimensional time domain matrix, and is constrained by tensor kernel norms; the interference signal has sparse property and tensor
Figure SMS_23
And (5) norm constraint.
In step 103, based on the tensor kernel norms and the tensors
Figure SMS_24
And constructing an objective function by using norms, constructing a constraint function based on the three-dimensional radar signals, the objective signals and the interference signals, and constructing an interference suppression model according to the objective function and the constraint function.
Illustratively, the target signal has a special edge channel low rank property on the three-dimensional time domain matrix, and the interference signal has a sparse property. Based on these two features, an interference suppression model is proposed to eliminate the effect of interference on useful target echoes. Specifically, the interference suppression model includes a solvable objective function and a constraint function.
In step 104, the interference suppression model is solved by an alternate direction multiplier method, so as to obtain a real target signal after interference suppression.
Illustratively, the above-described optimization problem is solved by an alternate direction multiplier method, in which, at the time of solving, three-dimensional data needs to be converted into two-dimensional data of different channels, and the solution is performed for each channel, to obtain the result of interference suppression.
Further, the tensor-based kernel norms and the tensors
Figure SMS_25
The norm construction objective function includes:
the objective function is constructed according to the following equation:
Figure SMS_26
，
wherein,,
Figure SMS_27
representing the target signal,/->
Figure SMS_28
Representing the interference signal,/->
Figure SMS_29
Representing a first superparameter,/->
Figure SMS_30
Representing tensor kernel norms,/->
Figure SMS_31
Representing tensor->
Figure SMS_32
Norms (F/F)>
Figure SMS_33
Indicating that the parameters in brackets are minimized.
Illustratively, in the above formula,
Figure SMS_34
tensor kernel norms representing tensors of the target signal,/->
Figure SMS_35
Tensor +.>
Figure SMS_36
Norms.
Further, the constructing a constraint function based on the three-dimensional radar signal, the target signal, and the interference signal includes:
constructing the constraint function according to the following formula:
Figure SMS_37
，
wherein,,
Figure SMS_38
representing the three-dimensional radar signal,/->
Figure SMS_39
Representing the target signal,/->
Figure SMS_40
Representing the interference signal,/->
Figure SMS_41
Represents noise margin, ++>
Figure SMS_42
The Frobenius norm of the tensor is represented.
By way of example only, and in an illustrative,
Figure SMS_43
abbreviations representing subject to +.>
Figure SMS_44
Frobenius norms representing tensors based on three-dimensional radar signals>
Figure SMS_45
Target signal->
Figure SMS_47
Interference signal->
Figure SMS_48
And constructing a constraint function.
Further, the solving the interference suppression model by the alternate direction multiplier method includes:
constructing an augmented Lagrangian function according to the objective function and the constraint function;
and constructing a threshold function and a soft threshold function, and solving the augmented Lagrangian function by an alternate direction multiplier method.
Further, the constructing an augmented lagrangian function from the objective function and the constraint function includes:
and constructing the augmented Lagrange function according to the objective function and the constraint function by taking the objective signal, the interference signal, the Lagrange variable and the second super parameter as variables.
Illustratively, the augmented Lagrangian function is an optimization algorithm, and the parameters to be solved are the variables of the augmented Lagrangian function.
Further, the constructing a threshold function and a soft threshold function, and solving the augmented lagrangian function by an alternating direction multiplier method includes:
decomposing the three-dimensional radar signals into two-dimensional radar signals corresponding to the channels;
performing multiple iterations on the two-dimensional radar signals of each channel, and calculating an updated two-dimensional target signal closed solution according to a singular value decomposition method and a threshold function in each iteration;
performing inverse Fourier transform on a plurality of two-dimensional target signal closed solutions corresponding to the channels to obtain a target signal three-dimensional tensor;
calculating a two-dimensional interference signal closed solution according to the soft threshold function and each two-dimensional target signal closed solution;
obtaining a three-dimensional tensor of the interference signal according to closed solution of a plurality of two-dimensional interference signals corresponding to the channels;
obtaining a two-dimensional Lagrangian variable closed solution according to the two-dimensional target signal closed solution and the two-dimensional interference signal closed solution;
obtaining a Lagrange variable three-dimensional tensor according to a plurality of two-dimensional Lagrange variable closed solutions corresponding to the channels;
calculating a super-parameter closed solution and calculating a termination condition function;
and stopping iteration when the iteration number reaches the preset iteration number or the value of the termination condition function is smaller than the preset error tolerance.
For example, by performing multiple iterations through the alternate direction multiplier method, a soft threshold function and a norm threshold function need to be constructed in the calculation process, and after the multiple iterations, the iterations stop when a preset condition is met.
Further, the performing multiple iterations on the two-dimensional radar signal of each channel, and calculating an updated two-dimensional target signal closed-form solution according to the singular value decomposition method and the threshold function in each iteration includes:
performing Fourier transformation on a target signal in the three-dimensional radar signal to obtain a preprocessed target signal, and obtaining a two-dimensional target signal of each channel according to the preprocessed target signal;
based on the two-dimensional radar signals of each channel and the two-dimensional Lagrange variable closed solution and the two-dimensional interference signal closed Jie Shengcheng first matrix obtained by the last iteration of the channel;
singular value decomposition is carried out on the first matrix to obtain a left singular matrix, a two-dimensional target signal closed solution and a right singular matrix;
and generating an updated two-dimensional target signal closed solution based on the left singular matrix, the two-dimensional target signal closed solution, the right singular matrix and the threshold function.
Further, the calculating the termination condition function includes:
and calculating the termination condition function based on the three-dimensional radar signal, the target signal three-dimensional tensor obtained after iteration, the interference signal three-dimensional tensor, the Lagrangian variable three-dimensional tensor and the super-parameter closed solution.
By way of example, the interference suppression model is solved by the alternative direction multiplier method through specific steps and formulas, and a real target signal after interference suppression is obtained. Specifically, the first step is obtained according to the following steps
Figure SMS_49
Each variable closed-form solution updated alternately for a plurality of iterations:
(1) Performing Fourier transformation on a target signal in the three-dimensional radar signal to obtain a preprocessed target signal, and obtaining a two-dimensional target signal of each channel according to the preprocessed target signal;
the pre-processing target signal at the t iteration is obtained according to the following formula:
Figure SMS_50
，
wherein,,
Figure SMS_51
representing the pre-processing target signal, FFT () representing a Fourier transform function, +>
Figure SMS_52
Representing the three-dimensional radar signal, 3 represents the 3 rd dimension data, i.e. the channel dimension.
(2) Based on the two-dimensional radar signals of each channel and the two-dimensional Lagrange variable closed solution and the two-dimensional interference signal closed Jie Shengcheng first matrix obtained by the last iteration of the channel;
the first matrix at the t-th iteration corresponding to channel k is expressed by the following formula:
A
Figure SMS_53
，
wherein A represents the first matrix,
Figure SMS_54
representing a two-dimensional radar signal corresponding to channel k, +.>
Figure SMS_55
Closed solution of Lagrangian variable at t-th iteration corresponding to channel k is represented by +.>
Figure SMS_56
Representing a super-parametric closed-form solution at the t-th iteration,>
Figure SMS_57
and (5) representing a two-dimensional interference signal closed-form solution at the t-th iteration corresponding to the channel k.
(3) And carrying out singular value decomposition on the first matrix to obtain a left singular matrix, a two-dimensional target signal closed solution and a right singular matrix.
Singular value decomposition is performed on the data at the t-th iteration corresponding to the channel k according to the following formula:
Figure SMS_58
，
wherein,,
Figure SMS_59
left singular matrix at t-th iteration corresponding to channel k is represented, < >>
Figure SMS_60
Representing a closed-form solution of the two-dimensional target signal at the t-th iteration corresponding to channel k,/for the iteration>
Figure SMS_61
Right singular matrix at t-th iteration corresponding to channel k is represented, H represents matrix transposition, ++>
Figure SMS_62
Representing a two-dimensional radar signal corresponding to channel k, +.>
Figure SMS_63
Closed solution of Lagrangian variable at t-th iteration corresponding to channel k is represented by +.>
Figure SMS_64
Representing a super-parametric closed-form solution at the t-th iteration,>
Figure SMS_65
and (5) representing a two-dimensional interference signal closed-form solution at the t-th iteration corresponding to the channel k.
(4) Generating an updated two-dimensional target signal closed-form solution, namely a first two-dimensional target signal closed-form solution, based on the left singular matrix, the two-dimensional target signal closed-form solution, the right singular matrix and the threshold function
Figure SMS_66
And iterating and updating alternately two-dimensional target signal closed-form solutions.
Generating the first according to
Figure SMS_67
And (3) performing iterative updating on a two-dimensional target signal closed-form solution:
Figure SMS_68
，
wherein,,
Figure SMS_69
representing a closed solution of the two-dimensional target signal at the t+1st iteration corresponding to channel k, < ->
Figure SMS_70
Left singular matrix at t-th iteration corresponding to channel k is represented, < >>
Figure SMS_71
Representing a closed-form solution of the two-dimensional target signal at the t-th iteration corresponding to channel k,/for the iteration>
Figure SMS_72
Right singular matrix at t-th iteration corresponding to channel k is represented, H represents matrix transposition, ++>
Figure SMS_73
Representing the threshold function.
Wherein the threshold function
Figure SMS_74
The definition is as follows:
Figure SMS_75
，
wherein,,
Figure SMS_76
representing the threshold function->
Figure SMS_77
Representing a two-dimensional closed-form solution of the target signal at the t-th iteration corresponding to channel k, "q.q" represents the q-th row and q-th column of the matrix, max () represents the maximum value of the parameter in brackets,/>
Figure SMS_78
And the super-parametric closed-form solution at the t-th iteration is shown.
(5) And performing inverse Fourier transform on a plurality of two-dimensional target signal closed solutions corresponding to the channels to obtain a target signal three-dimensional tensor.
Two-dimensional object signal closed-form solution of each channel
Figure SMS_79
And further obtaining closed solutions corresponding to all channels.
Then, the target signal three-dimensional tensor at the t+1st iteration is obtained according to the following formula:
Figure SMS_80
，
wherein,,
Figure SMS_81
representing the three-dimensional tensor of the target signal at the t+1st iteration, IFFT () represents the inverse fourier transform function,/->
Figure SMS_82
Represents a plurality of two-dimensional target signal closed solutions corresponding to a plurality of channels at the t-th iteration, and 3 represents the 3 rd dimensionData, i.e., channel dimensions.
(6) Calculating a two-dimensional interference signal closed solution according to the soft threshold function and each two-dimensional target signal closed solution;
obtaining a three-dimensional tensor of the interference signal according to closed solution of a plurality of two-dimensional interference signals corresponding to the channels;
specifically, a two-dimensional interference signal closed-form solution at the t+1st iteration corresponding to the channel k is obtained according to the following formula:
Figure SMS_83
，
wherein,,
Figure SMS_84
representing a two-dimensional interference signal closed-form solution at the t+1st iteration corresponding to channel k, < ->
Figure SMS_85
Representing a soft threshold function->
Figure SMS_86
Representing a two-dimensional radar signal corresponding to channel k, +.>
Figure SMS_87
Closed solution of Lagrangian variable at t-th iteration corresponding to channel k is represented by +.>
Figure SMS_88
Representing a super-parametric closed-form solution at the t-th iteration,>
Figure SMS_89
representing a closed solution of the two-dimensional target signal at the t+1st iteration corresponding to channel k, < ->
Figure SMS_90
Representing a first hyper-parameter.
Wherein the soft threshold function
Figure SMS_91
The definition is as follows:
Figure SMS_92
，
wherein,,
Figure SMS_93
representing the threshold function->
Figure SMS_94
First dimension data representing a soft threshold function, which can be substituted in particular into the two-dimensional interference signal closed-form solution>
Figure SMS_95
"p.q" means the row p and column q of the matrix, max () means the maximum value of the parameter in brackets, +.>
Figure SMS_96
Representing preset parameters.
(7) Obtaining a two-dimensional Lagrangian variable closed solution according to the two-dimensional target signal closed solution and the two-dimensional interference signal closed solution;
obtaining a Lagrange variable three-dimensional tensor according to a plurality of two-dimensional Lagrange variable closed solutions corresponding to the channels;
specifically, a two-dimensional lagrangian variable closed solution at the t+1st iteration corresponding to the channel k is obtained according to the following formula:
Figure SMS_97
，
wherein,,
Figure SMS_98
represents the closed-form solution of the two-dimensional Lagrangian variable at the t+1st iteration corresponding to the channel k,
Figure SMS_99
closed solution of Lagrangian variable at t-th iteration corresponding to channel k is represented by +.>
Figure SMS_100
Representing a super-parametric closed-form solution at the t-th iteration,>
Figure SMS_101
representing a two-dimensional radar signal corresponding to channel k, +.>
Figure SMS_102
Representing a two-dimensional interference signal closed-form solution at the t+1st iteration corresponding to channel k, < ->
Figure SMS_103
And (3) representing a two-dimensional target signal closed-form solution at the t+1st iteration corresponding to the channel k.
(8) Calculating a super-parameter closed solution and calculating a termination condition function; and stopping iteration when the iteration number reaches the preset iteration number or the value of the termination condition function is smaller than the preset error tolerance.
The hyper-parametric closed-form solution at the t+1th iteration is calculated according to the following formula:
Figure SMS_104
，
wherein,,
Figure SMS_105
indicate->
Figure SMS_106
Hyper-variable closed-form solution at multiple iterations, +.>
Figure SMS_107
Indicate->
Figure SMS_108
Hyper-variable closed-form solution at multiple iterations, +.>
Figure SMS_109
Representing preset parameters.
When the number of iterations reaches the desired limit or the termination condition is met, the iterations may stop. The termination condition function is represented by the following formula:
Figure SMS_110
，
wherein,,
Figure SMS_111
representing three-dimensional radar signals, < >>
Figure SMS_118
Indicate->
Figure SMS_122
Lagrangian variable three-dimensional tensor at multiple iterations,>
Figure SMS_114
indicate->
Figure SMS_116
Super-parametric closed-form solution at multiple iterations, +.>
Figure SMS_121
Indicate->
Figure SMS_124
Interference signal three-dimensional tensor in multiple iterations, < >>
Figure SMS_112
Indicate->
Figure SMS_115
Target signal three-dimensional tensor at multiple iterations, < >>
Figure SMS_119
Frobenius norms representing tensors, < ->
Figure SMS_123
，/>
Figure SMS_113
Represents the relative tolerance, wherein->
Figure SMS_117
The fast-time dimension of the tensor is represented,
Figure SMS_120
representing the slow time dimension of the tensor.
And finally, obtaining a real target signal after interference suppression through TRPCA recovery. Fig. 3 is a range-doppler plot of a real target signal after interference suppression, after which a target signal with higher accuracy can be obtained.
Through one or more of the above embodiments of the present invention, at least the following technical effects can be achieved:
in the technical scheme disclosed by the invention, the special edge channel low-rank property of the useful target signal on the three-dimensional time domain matrix and the sparse property of the interference signal are analyzed, and the interference suppression problem is constructed to separate the interference signal with sparse property and the target signal with low-rank property. And solving an interference suppression model through an Alternating Direction Multiplier Method (ADMM) to obtain an interference suppression result. The interference suppression method is suitable for various application scenes with longer interference duration and more interference numbers, and compared with the traditional method, the accuracy of suppressing the interference signals can be obviously improved.
Fig. 4 is a schematic structural diagram of a radar interference suppression device according to an embodiment of the present invention. According to another aspect of the present invention, there is also provided a radar interference suppression device for a MIMO radar, where the MIMO radar includes a plurality of transmitting antennas and a plurality of receiving antennas, the plurality of transmitting antennas and the plurality of receiving antennas form a plurality of channels, and referring to fig. 4, the device includes:
the signal acquisition module 201 is configured to acquire radar echo signals obtained by the radar in the multiple channels, and construct three-dimensional radar signals in a slow time dimension, a fast time dimension and a channel dimension according to the radar echo signals;
a norm constraint module 202 for constraining a target signal in the three-dimensional radar signal with a tensor kernel norm, using a tensor
Figure SMS_125
Norms constrain interfering signals in the three-dimensional radar signal;
a model construction module 203 for constructing a model based on the tensor kernel norms and the tensors
Figure SMS_126
Establishing an objective function by using norms, establishing a constraint function based on the three-dimensional radar signals, the objective signals and the interference signals, and establishing an interference suppression model according to the objective function and the constraint function;
and the calculation module 204 is configured to solve the interference suppression model by using an alternate direction multiplier method, so as to obtain a real target signal after interference suppression.
Further, the model building module 203 is further configured to:
the objective function is constructed according to the following equation:
Figure SMS_127
，
wherein,,
Figure SMS_128
representing the target signal,/->
Figure SMS_129
Representing the interference signal,/->
Figure SMS_130
Representing a first superparameter,/->
Figure SMS_131
Representing tensor kernel norms,/->
Figure SMS_132
Representing tensor->
Figure SMS_133
Norms (F/F)>
Figure SMS_134
Indicating that the parameters in brackets are minimized.
Further, the model building module 203 is further configured to:
constructing the constraint function according to the following formula:
Figure SMS_135
，
wherein,,
Figure SMS_136
representing the three-dimensional radar signal,/->
Figure SMS_137
Representing the target signal,/->
Figure SMS_138
Representing the interference signal,/->
Figure SMS_139
Represents noise margin, ++>
Figure SMS_140
The Frobenius norm of the tensor is represented.
Further, the computing module 204 is further configured to:
constructing an augmented Lagrangian function according to the objective function and the constraint function;
and constructing a threshold function and a soft threshold function, and solving the augmented Lagrangian function by an alternate direction multiplier method.
Further, the computing module 204 is further configured to:
and constructing the augmented Lagrange function according to the objective function and the constraint function by taking the objective signal, the interference signal, the Lagrange variable and the second super parameter as variables.
Further, the computing module 204 is further configured to:
decomposing the three-dimensional radar signals into two-dimensional radar signals corresponding to the channels;
performing multiple iterations on the two-dimensional radar signals of each channel, and calculating an updated two-dimensional target signal closed solution according to a singular value decomposition method and a threshold function in each iteration;
performing inverse Fourier transform on a plurality of two-dimensional target signal closed solutions corresponding to the channels to obtain a target signal three-dimensional tensor;
calculating a two-dimensional interference signal closed solution according to the soft threshold function and each two-dimensional target signal closed solution;
obtaining a three-dimensional tensor of the interference signal according to closed solution of a plurality of two-dimensional interference signals corresponding to the channels;
obtaining a two-dimensional Lagrangian variable closed solution according to the two-dimensional target signal closed solution and the two-dimensional interference signal closed solution;
obtaining a Lagrange variable three-dimensional tensor according to a plurality of two-dimensional Lagrange variable closed solutions corresponding to the channels;
calculating a super-parameter closed solution and calculating a termination condition function;
and stopping iteration when the iteration number reaches the preset iteration number or the value of the termination condition function is smaller than the preset error tolerance.
Further, the computing module 204 is further configured to:
performing Fourier transformation on a target signal in the three-dimensional radar signal to obtain a preprocessed target signal, and obtaining a two-dimensional target signal of each channel according to the preprocessed target signal;
based on a two-dimensional radar signal of each channel and a two-dimensional Lagrangian variable closed solution and a two-dimensional interference signal closed Jie Shengcheng first matrix obtained by the last iteration of the channel;
singular value decomposition is carried out on the first matrix to obtain a left singular matrix, a two-dimensional target signal closed solution and a right singular matrix;
and generating an updated two-dimensional target signal closed solution based on the left singular matrix, the two-dimensional target signal closed solution, the right singular matrix and the threshold function.
Further, the computing module 204 is further configured to:
and calculating the termination condition function based on the three-dimensional radar signal, the target signal three-dimensional tensor obtained after iteration, the interference signal three-dimensional tensor, the Lagrangian variable three-dimensional tensor and the super-parameter closed solution.
Other aspects and implementation details of the radar interference suppression device are the same as or similar to those of the radar interference suppression method described above, and are not described herein.
According to another aspect of the present invention, there is also provided a storage medium having stored therein a plurality of instructions adapted to be loaded by a processor to perform any of the radar disturbance suppression methods as described above.
In summary, although the present invention has been described in terms of the preferred embodiments, the preferred embodiments are not limited to the above embodiments, and various modifications and changes can be made by one skilled in the art without departing from the spirit and scope of the invention, and the scope of the invention is defined by the appended claims.

Claims (8)

1. A method for suppressing radar interference, for a MIMO vehicle-mounted millimeter wave radar, the MIMO vehicle-mounted millimeter wave radar including a plurality of transmitting antennas and a plurality of receiving antennas, the plurality of transmitting antennas and the plurality of receiving antennas forming a plurality of channels, the method comprising:
acquiring radar echo signals obtained by the radar in the multiple channels, and constructing three-dimensional radar signals in a slow time dimension, a fast time dimension and a channel dimension according to the radar echo signals;
constraining a target signal in the three-dimensional radar signal with a tensor kernel norm, using a tensor
Figure QLYQS_1
Norms constrain interfering signals in the three-dimensional radar signal;
based on the tensor kernel norms and the tensor
Figure QLYQS_2
Establishing an objective function by using norms, establishing a constraint function based on the three-dimensional radar signals, the objective signals and the interference signals, and establishing an interference suppression model according to the objective function and the constraint function;
solving the interference suppression model by an alternate direction multiplier method to obtain a real target signal after interference suppression;
wherein said solving the interference suppression model by the alternate direction multiplier method comprises:
constructing an augmented Lagrangian function according to the objective function and the constraint function;
constructing a threshold function and a soft threshold function, and solving the augmented Lagrangian function by an alternate direction multiplier method;
wherein constructing the threshold function and the soft threshold function and solving the augmented lagrangian function by an alternating direction multiplier method comprises:
decomposing the three-dimensional radar signals into two-dimensional radar signals corresponding to the channels;
performing multiple iterations on the two-dimensional radar signals of each channel, and calculating an updated two-dimensional target signal closed solution according to a singular value decomposition method and a threshold function in each iteration;
performing inverse Fourier transform on a plurality of two-dimensional target signal closed solutions corresponding to the channels to obtain a target signal three-dimensional tensor;
calculating a two-dimensional interference signal closed solution according to the soft threshold function and each two-dimensional target signal closed solution;
obtaining a three-dimensional tensor of the interference signal according to closed solution of a plurality of two-dimensional interference signals corresponding to the channels;
obtaining a two-dimensional Lagrangian variable closed solution according to the two-dimensional target signal closed solution and the two-dimensional interference signal closed solution;
obtaining a Lagrange variable three-dimensional tensor according to a plurality of two-dimensional Lagrange variable closed solutions corresponding to the channels;
calculating a super-parameter closed solution and calculating a termination condition function;
and stopping iteration when the iteration number reaches the preset iteration number or the value of the termination condition function is smaller than the preset error tolerance.
2. The suppression method of claim 1, wherein the tensor kernel norms and the tensors are based on
Figure QLYQS_3
The norm construction objective function includes:
the objective function is constructed according to the following equation:
Figure QLYQS_4
，
wherein,,
Figure QLYQS_5
representing the target signal,/->
Figure QLYQS_6
Representing the interference signal,/->
Figure QLYQS_7
Representing a first superparameter,/->
Figure QLYQS_8
Representing tensor kernel norms,/->
Figure QLYQS_9
Representing tensor->
Figure QLYQS_10
Norms (F/F)>
Figure QLYQS_11
Indicating that the parameters in brackets are minimized.
3. The suppression method of claim 2, wherein the constructing a constraint function based on the three-dimensional radar signal, the target signal, and the interference signal comprises:
constructing the constraint function according to the following formula:
Figure QLYQS_12
，
wherein,,
Figure QLYQS_13
representing the three-dimensional radar signal,/->
Figure QLYQS_14
Representing the target signal,/->
Figure QLYQS_15
Representing the interference signal,/->
Figure QLYQS_16
Represents noise margin, ++>
Figure QLYQS_17
The Frobenius norm of the tensor is represented.
4. The suppression method of claim 1, wherein the constructing an augmented lagrangian function from the objective function and the constraint function comprises:
and constructing the augmented Lagrange function according to the objective function and the constraint function by taking the objective signal, the interference signal, the Lagrange variable and the second super parameter as variables.
5. The suppression method of claim 4, wherein the two-dimensional radar signal for each channel is iterated a plurality of times, and in each iteration, calculating an updated two-dimensional target signal closed-form solution according to a singular value decomposition method and a threshold function comprises:
performing Fourier transformation on a target signal in the three-dimensional radar signal to obtain a preprocessed target signal, and obtaining a two-dimensional target signal of each channel according to the preprocessed target signal;
based on the two-dimensional radar signals of each channel and the two-dimensional Lagrange variable closed solution and the two-dimensional interference signal closed Jie Shengcheng first matrix obtained by the last iteration of the channel;
singular value decomposition is carried out on the first matrix to obtain a left singular matrix, a two-dimensional target signal closed solution and a right singular matrix;
and generating an updated two-dimensional target signal closed solution based on the left singular matrix, the two-dimensional target signal closed solution, the right singular matrix and the threshold function.
6. The suppression method of claim 5, wherein the calculating a termination condition function includes:
and calculating the termination condition function based on the three-dimensional radar signal, the target signal three-dimensional tensor obtained after iteration, the interference signal three-dimensional tensor, the Lagrangian variable three-dimensional tensor and the super-parameter closed solution.
7. A suppression device of radar interference for a MIMO vehicle-mounted millimeter wave radar, the MIMO vehicle-mounted millimeter wave radar including a plurality of transmitting antennas and a plurality of receiving antennas, the plurality of transmitting antennas and the plurality of receiving antennas constituting a plurality of channels, the device comprising:
the signal acquisition module is used for acquiring radar echo signals obtained by the radar in the multiple channels and constructing three-dimensional radar signals in a slow time dimension, a fast time dimension and a channel dimension according to the radar echo signals;
a norm constraint module for constraining the target signal in the three-dimensional radar signal by tensor kernel norms by tensor
Figure QLYQS_18
Norms constrain interfering signals in the three-dimensional radar signal;
a model construction module for constructing a model based on the tensor kernel norms and the tensors
Figure QLYQS_19
Establishing an objective function by using norms, establishing a constraint function based on the three-dimensional radar signals, the objective signals and the interference signals, and establishing an interference suppression model according to the objective function and the constraint function;
the calculation module is used for solving the interference suppression model through an alternate direction multiplier method so as to obtain a real target signal after interference suppression;
wherein the computing module is further configured to:
constructing an augmented Lagrangian function according to the objective function and the constraint function;
constructing a threshold function and a soft threshold function, and solving the augmented Lagrangian function by an alternate direction multiplier method;
wherein the computing module is further configured to:
decomposing the three-dimensional radar signals into two-dimensional radar signals corresponding to the channels;
performing multiple iterations on the two-dimensional radar signals of each channel, and calculating an updated two-dimensional target signal closed solution according to a singular value decomposition method and a threshold function in each iteration;
performing inverse Fourier transform on a plurality of two-dimensional target signal closed solutions corresponding to the channels to obtain a target signal three-dimensional tensor;
calculating a two-dimensional interference signal closed solution according to the soft threshold function and each two-dimensional target signal closed solution;
obtaining a three-dimensional tensor of the interference signal according to closed solution of a plurality of two-dimensional interference signals corresponding to the channels;
obtaining a two-dimensional Lagrangian variable closed solution according to the two-dimensional target signal closed solution and the two-dimensional interference signal closed solution;
obtaining a Lagrange variable three-dimensional tensor according to a plurality of two-dimensional Lagrange variable closed solutions corresponding to the channels;
calculating a super-parameter closed solution and calculating a termination condition function;
and stopping iteration when the iteration number reaches the preset iteration number or the value of the termination condition function is smaller than the preset error tolerance.
8. A storage medium having stored therein a plurality of instructions adapted to be loaded by a processor to perform the suppression method according to any one of claims 1 to 6.
CN202310433569.6A 2023-04-21 2023-04-21 Radar interference suppression method, device and storage medium Active CN116184332B (en)

Priority Applications (1)

Application Number Priority Date Filing Date Title
CN202310433569.6A CN116184332B (en) 2023-04-21 2023-04-21 Radar interference suppression method, device and storage medium

Applications Claiming Priority (1)

Application Number Priority Date Filing Date Title
CN202310433569.6A CN116184332B (en) 2023-04-21 2023-04-21 Radar interference suppression method, device and storage medium

Publications (2)

Publication Number Publication Date
CN116184332A CN116184332A (en) 2023-05-30
CN116184332B true CN116184332B (en) 2023-07-11

Family

ID=86434793

Family Applications (1)

Application Number Title Priority Date Filing Date
CN202310433569.6A Active CN116184332B (en) 2023-04-21 2023-04-21 Radar interference suppression method, device and storage medium

Country Status (1)

Country Link
CN (1) CN116184332B (en)

Family Cites Families (6)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US11907829B2 (en) * 2019-03-14 2024-02-20 Infineon Technologies Ag FMCW radar with interference signal suppression using artificial neural network
CN111610522B (en) * 2020-06-04 2021-12-24 中国人民解放军国防科技大学 SA-ISAR imaging method for target with micro-motion component based on low-rank and sparse combined constraint
CN111812648B (en) * 2020-07-22 2022-01-04 东南大学 Multichannel synthetic aperture radar RPCA amplitude-phase combined target detection method and device
CN114844544B (en) * 2022-04-28 2024-05-14 中国人民解放军国防科技大学 Mutual mass array beam forming method, system and medium based on low-rank tensor decomposition
CN115587281A (en) * 2022-09-29 2023-01-10 南京信息工程大学 Array element failure MIMO radar angle estimation method based on factor matrix prior
CN115963457B (en) * 2023-03-16 2023-07-25 东南大学 Multichannel passive radar direct wave interference suppression method

Also Published As

Publication number Publication date
CN116184332A (en) 2023-05-30

Similar Documents

Publication Publication Date Title
CN111142105B (en) ISAR imaging method for complex moving target
US11719811B2 (en) Method and device to improve radar data using reference data
US11506776B2 (en) Method and device with improved radar resolution
CN110568434B (en) Multi-channel uniform acceleration SAR moving target two-dimensional speed estimation method
CN112731307B (en) RATM-CFAR detector based on distance-angle joint estimation and detection method
CN102914773B (en) Multi-pass circumference SAR three-dimensional imaging method
Ran et al. An adaptive fast factorized back-projection algorithm with integrated target detection technique for high-resolution and high-squint spotlight SAR imagery
CN109669182B (en) Passive bistatic SAR moving/static target joint sparse imaging method
CN105223557B (en) Airborne early warning radar clutter suppression method based on accessory channel
CN101493934A (en) Weak target detecting method based on generalized S-transform
CN111551922A (en) Three-dimensional space double/multi-base radar high-speed target detection method
US11789142B2 (en) Graph-based array signal denoising for perturbed synthetic aperture radar
Abu et al. CFAR detection algorithm for objects in sonar images
CN108196238B (en) Clutter map detection method based on adaptive matched filtering under Gaussian background
CN112462367B (en) Vehicle detection method based on polarized synthetic aperture radar
CN116184332B (en) Radar interference suppression method, device and storage medium
CN117368877A (en) Radar image clutter suppression and target detection method based on generation countermeasure learning
US20220242443A1 (en) Method for detecting objects in automotive-grade radar signals
Li et al. Flexibility-residual BiSeNetV2 for GPR image decluttering
CN115561728A (en) Single-frame speed deblurring method and device of FMCW radar
CN115453523A (en) Scanning radar sparse target batch processing super-resolution method
CN112689773B (en) Radar signal processing method and radar signal processing device
CN116381617A (en) Radar interference suppression method, radar interference suppression device and storage medium
CN111880154A (en) Complex image domain moving object detection method based on symmetrical wave number spectrum cancellation
CN116559793A (en) Radar interference suppression method, radar interference suppression device and storage medium

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
GR01 Patent grant
GR01 Patent grant