MIMO millimeter wave radar speed angle joint estimation method and device
Through the combined velocity angle estimation method of MIMO millimeter wave radar, the recursive inversion technology and 2 FFT processing are used to solve the problems of low resolution and poor robustness in the traditional method, achieving higher resolution performance and lower computational complexity , obtain more accurate target speed and angle information.
Patent Information
- Application Number
- CN202410964595.6
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- Filing Date
- 2024-07-18
- Publication Date
- 2024-11-08
AI Technical Summary
Traditional MIMO millimeter wave radars have problems of low resolution and poor robustness in estimating motion target parameters, making it difficult to effectively obtain accurate speed and angle information.
A joint estimation method for MIMO millimeter wave radar velocity angle is proposed. By obtaining the sampled data cube construction velocity-angle joint data, using the recursive inversion technology of the sampling covariance matrix, combining the conditioning factors α and γ, the calculation complexity is reduced, and the calculation complexity is reduced through Two FFT processing achieves joint estimation of speed-angle.
It improves the resolution performance of speed and angle, enhances the robustness of the algorithm, and effectively reduces the computational complexity, and the obtained target spectrum peak is sharper, has a lower noise floor and reduces the running time by half.
Smart Images

Figure CN118915045A_ABST
Abstract
Description
Technical Field
[0001] The present invention relates to the field of radar related technologies, and in particular to a MIMO millimeter wave radar velocity angle joint estimation method and device. Background Art
[0002] Because it can effectively expand the virtual antenna aperture and thus improve angular resolution, MIMO (multiple-input-multiple-output) millimeter-wave radar has been widely used in smart driving, smart home, smart transportation and other fields.
[0003] In these applications, moving target parameter estimation is one of the most important functions of the millimeter wave radar system, and its performance is a key technical indicator for measuring the radar system. Traditionally, by performing three sequential FFT processes on the data cube (sampling values in the three dimensions of fast time, slow time, and space), the distance, speed, and angle of the moving target can be obtained. However, this processing method has problems such as low resolution and poor robustness. Summary of the invention
[0004] The purpose of the present invention is to solve at least one of the deficiencies of the prior art and to provide a MIMO millimeter-wave radar velocity angle joint estimation method and device.
[0005] In order to achieve the above object, the present invention adopts the following technical solutions:
[0006] Specifically, a MIMO millimeter-wave radar velocity angle joint estimation method is proposed, including the following:
[0007] Acquire a sampling data cube of the MIMO radar, and construct velocity-angle joint data based on the sampling data cube;
[0008] The velocity-angle joint data is processed by recursive calculation of sampling covariance matrix inversion to obtain the inverse matrix;
[0009] Based on the inverse matrix, a single velocity-angle joint spectrum corresponding to each column vector of the inverse matrix is obtained;
[0010] The final velocity-angle joint spectrum is obtained based on the single velocity-angle joint spectrum corresponding to each column vector of the inverse matrix, thereby realizing the velocity-angle joint estimation of the MIMO millimeter-wave radar.
[0011] Further, specifically, constructing the speed-angle joint data based on the sampling data cube includes:
[0012] The sampling data cube is recorded as D[M, N, T], where M, N, and T are the number of samples in the spatial dimension, slow time dimension, and fast time dimension, respectively, that is, these three variables correspond to the number of virtual array element channels, the number of chirps within the coherent accumulation time, and the number of samples on each chirp, respectively; the sampling data cube D[MN, T] is deformed to obtain the velocity-angle joint data Y[MN, T]. Each column of data comes from the Kronecker product of spatial dimension data and slow time dimension data, where Represents the complex number field.
[0013] Further, specifically, the velocity-angle joint data is processed by recursive calculation of the sampling covariance matrix inversion to obtain an inverse matrix, including,
[0014] Let the i-th column of Y be the column vector y i , i = 1, 2, ..., T, then the sampling covariance of the velocity-angle joint data is expressed as:
[0015]
[0016] in(·) H Represents the conjugate transpose of the matrix. For this reason, the present invention constructs a recursive inversion technique to reduce the computational burden. Considering the existence of noise and clutter in the sampled data, when recursively calculating the inverse matrix, the adjustment factors α (0<α<<1) and γ (0<γ<1) are added to make the velocity-angle joint estimation more robust.
[0017] Let I denote the unit matrix of MN×MN, β = 1 / α,
[0018] The constructed matrix recursive inversion technique is obtained using the following loop:
[0019] for i=2:T
[0020] b=R 1 y i ;
[0021]
[0022] R 1 =R i ;
[0023] end
[0024] Among them, α and γ are adjustment factors, and when setting, they satisfy 0<α<<1, 0<γ<1.
[0025] The final inverse matrix is recorded as Let its kth column be a vector k=1,2,…,MN
[0026] Further, specifically, based on the inverse matrix, a single velocity-angle joint spectrum corresponding to each column vector of the inverse matrix is obtained, including.
[0027] For the kth column vector of the inverse matrix,
[0028] First, r k Rearrange into a matrix The rearrangement here is equivalent to r k The 1 to N rows are G k The first column and rows N+1 to 2N are G k The second column of , and so on for the rest;
[0029] Secondly, for the matrix G k Perform FFT processing on each column and record the result as V k ;
[0030] Again, for V k Perform FFT processing on each row of k ;
[0031] Finally, a single velocity-angle joint spectrum J is obtained k =1. / |A k |, where |·| represents the modulo operation, 1. / |A k |Represents the pair|A k |Element-wise reciprocal;
[0032] For each k=1, 2, ..., MN, the above steps are performed in sequence to obtain a single velocity-angle joint spectrum corresponding to each column vector of the inverse matrix.
[0033] Further, specifically, the final velocity-angle joint spectrum is obtained based on the single velocity-angle joint spectrum corresponding to each column vector of the inverse matrix, including:
[0034] The single velocity-angle joint spectrum corresponding to each column vector of the inverse matrix is averaged to obtain the final velocity-angle joint spectrum. The peak value of J corresponds to the target of interest, and the peak coordinate value represents the speed and angle information of the target.
[0035] The present invention also proposes a MIMO millimeter wave radar velocity angle joint estimation device, comprising:
[0036] A data acquisition module, used for acquiring a sampling data cube of the MIMO radar, and constructing velocity-angle joint data based on the sampling data cube;
[0037] An inverse matrix calculation module is used to process the velocity-angle joint data by recursive calculation of the sampling covariance matrix inversion to obtain an inverse matrix;
[0038] A velocity-angle joint spectrum calculation module, used for obtaining a single velocity-angle joint spectrum corresponding to each column vector of the inverse matrix based on the inverse matrix;
[0039] The speed-angle joint estimation module is used to obtain the final speed-angle joint spectrum based on the single speed-angle joint spectrum corresponding to each column vector of the inverse matrix, thereby realizing the speed-angle joint estimation of the MIMO millimeter-wave radar.
[0040] The beneficial effects of the present invention are:
[0041] The present invention proposes a MIMO millimeter-wave radar speed and angle joint estimation method and device, and proposes a parameter-adjusted matrix recursive inversion technology. On this basis, the two-dimensional problem of speed and angle joint estimation is converted into two one-dimensional FFTs, thereby realizing low-complexity speed-angle joint estimation. Compared with the prior art, the present invention has the following advantages and beneficial effects:
[0042] 1. Compared with the traditional matrix recursive inversion technology, the present invention introduces two adjustment parameters based on the characteristics of MIMO millimeter-wave radar data, thereby enhancing the robustness of the algorithm.
[0043] 2. Compared with the subspace-type super-resolution velocity-angle joint estimation method, the present invention realizes the velocity-angle joint estimation by rearranging data and using 2 FFTs based on the characteristics of the inverse matrix of the sampling covariance matrix, thus avoiding matrix-vector multiplication operations and effectively reducing the computational complexity of the algorithm. BRIEF DESCRIPTION OF THE DRAWINGS
[0044] The above and other features of the present disclosure will become more apparent by describing in detail the embodiments shown in the accompanying drawings. The same reference numerals in the accompanying drawings of the present disclosure represent the same or similar elements. Obviously, the accompanying drawings described below are only some embodiments of the present disclosure. For those skilled in the art, other accompanying drawings can be obtained based on these accompanying drawings without creative work. In the accompanying drawings:
[0045] Figure 1 It is a flow chart of a MIMO millimeter wave radar velocity angle joint estimation method of the present invention;
[0046] Figure 2 The figure shows a velocity-angle joint spectrum obtained when a MIMO millimeter-wave radar velocity angle joint estimation method of the present invention is applied;
[0047] Figure 3Shown is the velocity-angle joint spectrum obtained by the classic Capon method;
[0048] Figure 4 The target velocity-angle spectrum obtained by 2 FFTs is shown. DETAILED DESCRIPTION
[0049] The following will be combined with the embodiments and drawings to clearly and completely describe the concept, specific structure and technical effects of the present invention, so as to fully understand the purpose, scheme and effect of the present invention. It should be noted that the embodiments in this application and the features in the embodiments can be combined with each other without conflict. The same reference numerals used throughout the drawings indicate the same or similar parts.
[0050] Example 1, reference Figure 1 The present invention proposes a MIMO millimeter wave radar velocity angle joint estimation method, comprising the following steps:
[0051] Step 110: Obtain a sampling data cube of the MIMO radar, and construct velocity-angle joint data based on the sampling data cube;
[0052] Step 120, processing the velocity-angle joint data by recursive calculation of sampling covariance matrix inversion to obtain an inverse matrix;
[0053] Step 130: obtaining a single velocity-angle joint spectrum corresponding to each column vector of the inverse matrix based on the inverse matrix;
[0054] Step 140: Obtain a final velocity-angle joint spectrum based on a single velocity-angle joint spectrum corresponding to each column vector of the inverse matrix, thereby realizing velocity-angle joint estimation of the MIMO millimeter-wave radar.
[0055] In this embodiment 1, a MIMO millimeter-wave radar speed-angle joint estimation method is proposed, and a matrix recursive inversion technology with parameter adjustment is used. On this basis, the two-dimensional problem of joint speed and angle estimation is converted into two one-dimensional FFTs. This method does not require matching between angle and speed, and can effectively improve resolution performance and robustness, and reduce computational complexity.
[0056] As a preferred embodiment of the present invention, specifically, constructing velocity-angle joint data based on the sampling data cube includes:
[0057] The sampled data cube is recorded as D[M, N, T], where M, N, and T are the number of samples in the spatial dimension, slow time dimension, and fast time dimension, respectively. That is, these three variables correspond to the number of virtual array element channels, the number of chirps in the coherent process interval (CPI), and the number of samples on each chirp. The sampled data cube D[M, N, T] is deformed to obtain the velocity-angle joint data Y[MN, T], Each column of data comes from the Kronecker product of spatial dimension data and slow time dimension data, where C represents the complex number domain.
[0058] As a preferred embodiment of the present invention, specifically, the velocity-angle joint data is processed by recursive calculation of the sampling covariance matrix inversion to obtain an inverse matrix, including:
[0059] Let the i-th column of Y be the column vector y i , i = 1, 2, ..., T, then the sampling covariance of the velocity-angle joint data is expressed as:
[0060]
[0061] in(·) H Represents the conjugate transpose of the matrix. For this reason, the present invention constructs a recursive inversion technique to reduce the computational burden. Considering the existence of noise and clutter in the sampled data, when recursively calculating the inverse matrix, the adjustment factors α (0<α<<1) and γ (0<γ<1) are added to make the velocity-angle joint estimation more robust.
[0062] Let I denote the unit matrix of MN×MN, β = 1 / α,
[0063] The constructed matrix recursive inversion technique is obtained using the following loop:
[0064] fOr i=2:T
[0065] b=R 1 y i ;
[0066]
[0067] R 1 =R i ;
[0068] end
[0069] Among them, α and γ are adjustment factors, and when set, they satisfy 0<α<<1, 0<γ<1, and i=2: T means that i increases by 1 from 2 to T in sequence, and the calculation content of each cycle is,
[0070] Finally, the obtained inverse matrix is denoted as Let its k-th column be denoted as the vector k = 1, 2, …, MN
[0071] In this preferred embodiment, considering that the product of MN is usually very large, directly performing matrix inversion operation on R has a complexity as high as O(M 3 N 3 ), which is not conducive to implementation on an embedded system. Therefore, the present invention constructs a recursive inversion technique to reduce the computational burden. Considering that there is noise and clutter in the sampled data, when performing recursive calculation of the inverse matrix, adjustment factors α (0 < α << 1) and γ (0 < γ < 1) are added to make the speed-angle joint estimation more robust, and the complexity of the above-mentioned recursive inversion process proposed by the present invention is reduced to O(M 2 N 2 T). When T < MN, the complexity of the sampled covariance matrix inversion proposed by the present invention is lower than that of direct inversion.
[0072] As a preferred embodiment of the present invention, specifically, based on the inverse matrix, the single-shot speed-angle joint spectrum corresponding to each column vector of the inverse matrix is obtained, including,
[0073] For the k-th column vector of the inverse matrix,
[0074] First, rearrange r k into the matrix Here, the rearrangement is equivalent to taking the 1st to N-th rows of r k as the 1st column of G k , the (N + 1)-th to 2N-th rows as the 2nd column of G k , and so on;
[0075] Second, perform FFT processing on each column of the matrix G k , and denote the result as V k ;
[0076] Third, perform FFT processing on each row of V k , and denote the result as A k .
[0077] Finally, obtain the single-shot speed-angle joint spectrum J k = 1. / |A k |, where |·| represents the modulus operation, and 1. / |A k | represents taking the reciprocal of |A k | element by element;
[0078] For each k=1, 2, ..., MN, the above steps are performed in sequence to obtain a single velocity-angle joint spectrum corresponding to each column vector of the inverse matrix.
[0079] As a preferred embodiment of the present invention, specifically, the final velocity-angle joint spectrum is obtained based on the single velocity-angle joint spectrum corresponding to each column vector of the inverse matrix, including:
[0080] The single velocity-angle joint spectrum corresponding to each column vector of the inverse matrix is averaged to obtain the final velocity-angle joint spectrum. The peak value of J corresponds to the target of interest, and the peak coordinate value represents the speed and angle information of the target.
[0081] In the specific simulation verification,
[0082] The simulation conditions are as follows: Assume that the MIMO millimeter wave radar has a 1-dimensional equidistant linear array with 2 transmitters and 4 receivers, with a total of M = 8 virtual channels, and the array element spacing is half a wavelength. The number of samples in the fast time dimension is T = 512, and the number of samples in the slow time dimension is N = 128. There are two targets in the scene, whose signal-to-noise ratios (SNRs) are 10dB and 15dB respectively, the distances are 3.5m and 6.5m respectively, the speeds are 1.3m / s and 0.5m / s respectively, and the azimuths are -23.1° and 12.4° respectively. In the execution of the algorithm of the present invention, α = 10 -6 , γ=0.2, and the length of FFT during execution is 128. Figure 2 , Figure 3 , Figure 4 The target speed-angle joint spectrum obtained by the present invention, the speed-angle joint spectrum of the classical Capon method, and the target speed-angle spectrum obtained by the 2nd FFT are respectively given. Compared with the traditional 2nd FFT method, the target spectrum peak obtained by the method of the present invention is sharper (higher resolution) and has a lower noise floor. Compared with the classical Capon method, although the deviation of the target speed and angle values obtained by the present invention is slightly larger, the SNR is more in line with the actual situation between the two targets. In addition, in this implementation case, the running time of the proposed algorithm of the present invention is only 1 / 2 of that of the Capon method, which proves that the present invention can effectively reduce the computational complexity.
[0083] The present invention also proposes a MIMO millimeter wave radar velocity angle joint estimation device, comprising:
[0084] A data acquisition module, used for acquiring a sampling data cube of the MIMO radar, and constructing velocity-angle joint data based on the sampling data cube;
[0085] An inverse matrix calculation module is used to process the velocity-angle joint data by recursive calculation of the sampling covariance matrix inversion to obtain an inverse matrix;
[0086] A velocity-angle joint spectrum calculation module, used for obtaining a single velocity-angle joint spectrum corresponding to each column vector of the inverse matrix based on the inverse matrix;
[0087] The speed-angle joint estimation module is used to obtain the final speed-angle joint spectrum based on the single speed-angle joint spectrum corresponding to each column vector of the inverse matrix, thereby realizing the speed-angle joint estimation of the MIMO millimeter-wave radar.
[0088] In addition, each functional module in each embodiment of the present invention may be integrated into one processing module, or each module may exist physically separately, or two or more modules may be integrated into one module. The above-mentioned integrated module may be implemented in the form of hardware or in the form of software functional modules.
[0089] If the integrated module is implemented in the form of a software function module and sold or used as an independent product, it can be stored in a computer-readable storage medium. Based on this understanding, the present invention implements all or part of the processes in the above-mentioned embodiment method, and can also be completed by instructing the relevant hardware through a computer program. The computer program can be stored in a computer-readable storage medium, and the computer program can implement the steps of the above-mentioned various method embodiments when executed by the processor. Among them, the computer program includes computer program code, and the computer program code can be in source code form, object code form, executable file or some intermediate form. The computer-readable medium may include: any entity or system that can carry the computer program code, recording medium, U disk, mobile hard disk, disk, optical disk, computer memory, read-only memory (ROM, Read-Only Memory), random access memory (RAM, Random Access Memory), electrical carrier signal, telecommunication signal and software distribution medium, etc.
[0090] Although the description of the present invention has been quite detailed and specifically describes several described embodiments, it is not intended to be limited to any of these details or embodiments or any particular embodiment, but should be regarded as providing a broad possible interpretation of these claims in view of the prior art by reference to the appended claims, thereby effectively covering the intended scope of the present invention. In addition, the above description of the present invention is based on the embodiments foreseeable by the inventor, and its purpose is to provide a useful description, and those non-substantial changes to the present invention that have not yet been foreseen may still represent equivalent changes to the present invention.
[0091] The above is only a preferred embodiment of the present invention. The present invention is not limited to the above implementation. As long as the technical effect of the present invention is achieved by the same means, it should belong to the protection scope of the present invention. Within the protection scope of the present invention, its technical scheme and / or implementation method can have various modifications and changes.
Claims
1. A MIMO millimeter wave radar velocity angle joint estimation method, characterized in that: Includes the following: Acquire a sampling data cube of the MIMO radar, and construct velocity-angle joint data based on the sampling data cube; The velocity-angle joint data is processed by recursive calculation of sampling covariance matrix inversion to obtain the inverse matrix; Based on the inverse matrix, a single velocity-angle joint spectrum corresponding to each column vector of the inverse matrix is obtained; The final velocity-angle joint spectrum is obtained based on the single velocity-angle joint spectrum corresponding to each column vector of the inverse matrix, thereby realizing the velocity-angle joint estimation of the MIMO millimeter-wave radar.
2. A MIMO millimeter wave radar velocity angle joint estimation method according to claim 1, characterized in that: Specifically, constructing the velocity-angle joint data based on the sampling data cube includes: The sampled data cube is recorded as D[M, N, T], where M, N, and T are the number of samples in the spatial dimension, slow time dimension, and fast time dimension, respectively, that is, these three variables correspond to the number of virtual array element channels, the number of chirps in the coherent accumulation time, and the number of samples on each chirp, respectively; the sampled data cube D[M, N, T] is deformed to obtain the velocity-angle joint data Y[MN, T], Each column of data comes from the Kronecker product of spatial dimension data and slow time dimension data, where Represents the complex field.
3. The MIMO millimeter wave radar velocity angle joint estimation method according to claim 2, characterized in that: Specifically, the velocity-angle joint data is processed by recursive calculation of the sampling covariance matrix inversion to obtain the inverse matrix, include, Let the i-th column of Y be the column vector y i , i = 1, 2, ..., T, then the sampling covariance of the velocity-angle joint data is expressed as: in(·) H Represents the conjugate transpose of the matrix. For this reason, the present invention constructs a recursive inversion technique to reduce the computational burden. Considering the existence of noise and clutter in the sampled data, when recursively calculating the inverse matrix, the adjustment factors α (0 < α < < 1) and γ (0 < γ < 1) are added to make the velocity-angle joint estimation more robust. Let I denote the unit matrix of MN×MN, β = 1 / α, The constructed matrix recursive inversion technique is obtained using the following loop: for i=2:T b=R1y i ; R1=R i ; end Among them, α and γ are adjustment factors, and when setting, they satisfy 0<α<<1, 0<γ<1. Finally, the inverse matrix is recorded as Let its column be a vector 4. The MIMO millimeter wave radar velocity angle joint estimation method according to claim 3, characterized in that: Specifically, based on the inverse matrix, a single velocity-angle joint spectrum corresponding to each column vector of the inverse matrix is obtained, including: For the kth column vector of the inverse matrix, First, r k Rearrange into a matrix The rearrangement here is equivalent to r k The 1 to N rows are G k The first column and rows N+1 to 2N are G k The second column of , and so on for the rest; Secondly, for the matrix G k Perform FFT processing on each column and record the result as V k ; Again, for V k Perform FFT processing on each row of k ; Finally, a single velocity-angle joint spectrum J is obtained k =1. / |A k |, where |·| represents the modulo operation, 1. / |A k |Represents the pair|A k |Element-wise reciprocal; For each k=1, 2, ..., MN, the above steps are performed in sequence to obtain a single velocity-angle joint spectrum corresponding to each column vector of the inverse matrix.
5. A MIMO millimeter wave radar velocity angle joint estimation method according to claim 4, characterized in that: Specifically, the final velocity-angle joint spectrum is obtained based on the single velocity-angle joint spectrum corresponding to each column vector of the inverse matrix, including: The single velocity-angle joint spectrum corresponding to each column vector of the inverse matrix is averaged to obtain the final velocity-angle joint spectrum. The peak value of J corresponds to the target of interest, and the peak coordinate value represents the speed and angle information of the target.
6. A MIMO millimeter wave radar velocity angle joint estimation device, characterized in that: include: A data acquisition module, used for acquiring a sampling data cube of the MIMO radar, and constructing velocity-angle joint data based on the sampling data cube; An inverse matrix calculation module is used to process the velocity-angle joint data by recursive calculation of the sampling covariance matrix inversion to obtain an inverse matrix; A velocity-angle joint spectrum calculation module, used for obtaining a single velocity-angle joint spectrum corresponding to each column vector of the inverse matrix based on the inverse matrix; The speed-angle joint estimation module is used to obtain the final speed-angle joint spectrum based on the single speed-angle joint spectrum corresponding to each column vector of the inverse matrix, thereby realizing the speed-angle joint estimation of the MIMO millimeter-wave radar.
