Direct waveform design method, device, equipment and medium for MIMO transmission pattern

Abstract

The invention discloses a direct waveform design method, device, equipment and medium for a MIMO transmission pattern, and relates to the field of radar technology. The method comprises: obtaining a precoding matrix of a transmission signal and a steering vector of a transmission array, obtaining an initial transmission signal matrix vector quantizer, initial sidelobe area power data and initial mainlobe area power data based on the precoding matrix of the transmission signal and the steering vector of the transmission array, constructing an initial optimization model based on the obtained data, wherein the initial optimization model comprises an original objective function and an original constraint condition, wherein the original objective function is used to characterize the relationship between the initial sidelobe area power data and the initial mainlobe area power data, and the original constraint condition comprises a sidelobe constraint for limiting the sum of the sidelobe area power, a mainlobe constraint for limiting the sum of the mainlobe area power, and a total power constraint for limiting the transmission power of a transmitting antenna, wherein the model is processed by constructing auxiliary variables, and then ADMM is used to solve the model to realize the drawing of a target transmission pattern.

Description

Technical Field

[0001] The present invention relates to the field of radar technology, and in particular to a method, device, equipment and medium for direct waveform design of a MIMO transmission pattern. Background Art

[0002] In recent years, Multiple-Input Multiple-Output (MIMO) radar has attracted more and more attention from academia and industry due to its superior performance in target detection and parameter estimation. According to the array element structure, MIMO radar can be divided into distributed MIMO radar and centralized MIMO radar. MIMO radar has high flexibility in waveform design, which is reflected in the fact that each array element of MIMO radar can independently transmit waveforms. Therefore, how to use this feature to design different transmission waveforms according to specific target scenarios is a meaningful research topic.

[0003] Waveform optimization and transmission pattern matching are two key technologies of MIMO radar. Through reasonable waveform design, the generated transmission pattern can be as close to the expected transmission pattern as possible, which can achieve the goal of meeting actual needs. Among them, the waveform design of centralized MIMO radar is mainly divided into two categories: one is to improve the output signal-to-interference-noise ratio by jointly optimizing the transmission waveform and the receiving filter under interference conditions; the other is to use the centralized MIMO radar to control the energy distribution in space. By optimizing the transmission waveform, the transmission power is focused in the desired direction and the power of the side lobes is reduced. The transmission pattern matching based on centralized MIMO radar is mainly divided into two categories: one is to optimize the covariance matrix of the transmission waveform to obtain the target transmission pattern; the other is to directly establish the functional relationship between the transmission waveform and the target transmission pattern, and iterate the solution through the optimization algorithm. Since the covariance matrix is ​​a square matrix with a dimension equal to the number of transmitting array elements, the target transmission pattern is obtained by optimizing the covariance matrix of the transmission waveform. However, there are problems such as the inability to restore the waveform matrix by matrix decomposition and insufficient information. The currently commonly used method to directly establish the functional relationship between the transmission waveform and the target transmission pattern is the method of minimizing the template matching error, that is, taking the difference between the transmission pattern to be optimized and the known transmission pattern as the objective function to be solved. However, this method needs to solve a fourth-order non-convex optimization problem in the solution process, which has great difficulties in solving. In addition, since the method of minimizing the template matching error targets an entire airspace and lacks separate control of the main lobe or side lobe, the results are prone to problems such as excessive main lobe fluctuations and high side lobes. Summary of the invention

[0004] The technical problem to be solved by the present invention is that when the transmission pattern matching of a centralized MIMO radar is realized by minimizing the template matching error, it is difficult or even impossible to find a solution, and the result obtained by the solution is prone to problems such as excessive main lobe fluctuation and high side lobe, resulting in a large difference between the target transmission pattern and the expected transmission pattern. In order to solve this technical problem, the present invention provides a direct waveform design method, device, equipment and medium for a MIMO transmission pattern.

[0005] The technical solution of the present invention to solve the above technical problems is as follows:

[0006] A direct waveform design method for a MIMO transmission pattern, comprising:

[0007] Step S1: obtaining a precoding matrix of a transmitted signal and a steering vector of a transmitting array, obtaining an initial transmitted signal matrix vectorization operator and a covariance matrix of the steering vector of the transmitting array according to the precoding matrix of the transmitted signal and the steering vector of the transmitting array, and obtaining initial sidelobe area power data and initial mainlobe area power data according to the initial transmitted signal matrix vectorization operator and the covariance matrix;

[0008] Step S2: constructing an initial optimization model according to the initial transmission signal matrix vectorization operator, the initial sidelobe area power data and the initial mainlobe area power data, wherein the initial optimization model includes an original objective function and an original constraint condition, wherein the original objective function is used to characterize the relationship between the initial sidelobe area power data and the initial mainlobe area power data, and the original constraint condition includes a sidelobe constraint, a mainlobe constraint and a total power constraint, wherein the sidelobe constraint is used to limit the power sum in the sidelobe area, the mainlobe constraint is used to limit the power sum in the mainlobe area, and the total power constraint is used to limit the transmission power of the transmitting antenna;

[0009] Step S3: construct a first auxiliary variable, use the first auxiliary variable to simplify the original objective function to obtain a first intermediate function A, the first intermediate function A includes the maximum power data A in the sidelobe area and the minimum power data A in the mainlobe area, optimize the maximum power data A in the sidelobe area and the minimum power data A in the mainlobe area contained in the first intermediate function A to obtain a second intermediate function B, the second intermediate function B includes the maximum power data B in the sidelobe area and the minimum power data B in the mainlobe area, and construct a secondary optimization model according to the second intermediate function B and the original constraint conditions;

[0010] Step S4: introducing a second auxiliary variable, using the second auxiliary variable to transform the secondary optimization model to obtain a third intermediate function C and new constraints, wherein the third intermediate function C includes the maximum power data C1 in the sidelobe area and the minimum power data C2 in the mainlobe area; performing approximate processing on the maximum power data C1 in the sidelobe area and the minimum power data C2 in the mainlobe area contained in the third intermediate function C to obtain a new objective function G, and constructing a new optimization model according to the new objective function G and the new constraints;

[0011] Step S5: Use the alternating direction method of multipliers (ADMM) to solve the new optimization model to obtain the target transmission signal matrix vector operator, calculate the target main lobe area power and the target side lobe area power according to the target transmission signal matrix vector operator, and draw the target transmission direction diagram according to the target main lobe area power and the target side lobe area power.

[0012] The beneficial effects of the present invention are as follows: by establishing an initial optimization model, the power in the side lobe area and the main lobe area is optimized and constrained, thereby solving the problem that the traditional method lacks control over the main lobe or the side lobe, resulting in excessive main lobe fluctuation and high side lobe in the generated target emission pattern; by approximating the new optimization model, the fourth-order non-convex optimization problem in the new optimization model to be solved is transformed into a solvable second-order convex optimization problem, and then combined with the ADMM solution model, it is possible to achieve the generation of the required target emission pattern and achieve a high degree of matching between the target emission pattern and the expected emission pattern.

[0013] Based on the above technical solution, the present invention can also be improved as follows.

[0014] Further, the step S1 specifically includes:

[0015] The precoding matrix of the transmission signal of the centralized MIMO radar transmission array composed of multiple transmission array elements in the target direction θ and the steering vector of the transmission array are obtained, wherein the precoding matrix of the transmission signal is:

[0016] S=[S1,S2,...,S i ,...,S L ]

[0017] Wherein, S represents the precoding matrix of the transmitted signal, S i =[s1(l),s2(l),...,s m (l),...,s M (l)] T, i, m, l are all variables and their values ​​are positive integers, 1≤i≤L, 1≤m≤M, 1≤l≤L, L represents the signal code length, M represents the total number of transmitting array elements, s m (l) represents the transmission signal of the mth transmitting array element at time l, S i represents the transmission sequence of multiple transmitting array elements at time l;

[0018] The steering vector of the transmitting array is:

[0019]

[0020] Wherein, a(θ) represents the steering vector of the transmitting array, the steering vector of the transmitting array includes the steering vector of the side lobe area and the steering vector of the main lobe area, d represents the array element spacing, λ represents the signal wavelength, and j represents a unit imaginary number; (·) T Represents the transpose operation;

[0021] An initial transmit signal matrix vectorization operator is obtained according to the precoding matrix of the transmit signal, and the initial transmit signal matrix vectorization operator is:

[0022] s=vec(S)

[0023] Wherein, s represents the matrix vectorization operator of the initial transmission signal, and vec(·) represents the matrix straightening operation;

[0024] The covariance matrix of the steering vector of the transmit array is obtained according to the steering vector of the transmit array. The covariance matrix of the steering vector of the transmit array is:

[0025] R(θ)=a(θ)a H (θ)

[0026] Where R(θ) represents the covariance matrix of the steering vector of the transmit array, (·) H represents the conjugate transpose operation;

[0027] According to the initial transmission signal matrix vectorization operator and the covariance matrix, the initial sidelobe area power data is obtained, and the initial sidelobe area power data is Among them, Θ s Represents the side lobe area, and the power in the angle area corresponding to the side lobe area is θ s represents the sidelobe area angle variable and satisfies θ s ∈Θ s , represents the covariance matrix of the steering vectors in the sidelobe region;

[0028] According to the initial transmission signal matrix vectorization operator and the covariance matrix, the initial main lobe area power data is obtained, and the initial main lobe area power data is Among them, Θ z Represents the main lobe area, and the power in the angle area corresponding to the main lobe area is θ z represents the main lobe area angle variable, and satisfies θ s ∈Θ s , Represents the covariance matrix of the steering vectors in the mainlobe region.

[0029] Further, in step S2, constructing an initial optimization model according to the initial transmit signal matrix vector quantizer, the initial sidelobe area power data, and the initial mainlobe area power data specifically includes:

[0030] The original objective function is established according to the minimum ratio of the maximum power in the side lobe area to the minimum power in the main lobe area, and the original objective function is:

[0031]

[0032] A sidelobe constraint is established according to the power in the angle region corresponding to the sidelobe region, and the sidelobe constraint is:

[0033]

[0034] According to the power in the angle region corresponding to the main lobe region, a main lobe constraint is established, and the main lobe constraint is:

[0035]

[0036] According to the initial transmission signal matrix vectorization operator, a total power constraint is established, and the total power constraint is:

[0037] s2=P

[0038] Where σ1 represents the sum of the power in the sidelobe area, σ2 represents the sum of the power in the mainlobe area, and P represents the total power of the transmitting array element;

[0039] According to the original objective function, the side lobe constraint, the main lobe constraint, and the total power constraint, the initial optimization model is constructed, and the initial optimization model is:

[0040]

[0041] The beneficial effects of adopting the above further scheme are: the original objective function in the initial optimization model includes the peak power in the sidelobe area and the minimum power in the mainlobe area, which improves the pertinence of the design of the target transmission pattern; the precoding matrix of the transmitted signal is in the form of a vectorized operator of the waveform matrix, which avoids the problem that the target transmission pattern cannot be restored to the waveform matrix when solving the covariance matrix of the transmitted waveform; the original constraints of the initial optimization model include the total power constraints of the mainlobe area and the sidelobe area, which further improves the accuracy of the target transmission pattern matching.

[0042] Further, the step S3 specifically includes:

[0043] The first auxiliary variable g is constructed, and the initial sidelobe area power data and the initial mainlobe area power data in the original objective function are transformed by the first auxiliary variable g to obtain the first intermediate function A, where the first intermediate function A is:

[0044]

[0045] in, is the maximum power data A in the sidelobe area in the first intermediate function A, is the minimum power data A in the main lobe area in the first intermediate function A;

[0046] Based on the LES method and the exponential logarithmic function operation, the maximum power data A in the sidelobe area and the minimum power data A in the mainlobe area in the first intermediate function A are optimized to obtain the second intermediate function B, and the second intermediate function B is:

[0047]

[0048] in, α and ε are both positive constants, and I represents the identity matrix; is the maximum power data B in the sidelobe area in the second intermediate function B, is the minimum power data B in the main lobe area in the second intermediate function B;

[0049] According to the second intermediate function B and the original constraint condition, the secondary optimization model is constructed, and the secondary optimization model is:

[0050]

[0051] The beneficial effects of adopting the above-mentioned further scheme are: using the first auxiliary variable to simplify the initial sidelobe area power data and the initial mainlobe area power data in the original objective function, so as to convert the fraction of the original objective function into an integer, which is convenient for subsequent optimization; by optimizing the sidelobe area maximum power data A and the mainlobe area minimum power data A in the first intermediate function A, the consistency of the sidelobe area maximum power data A and the mainlobe area minimum power data A and the positive definiteness of the covariance matrix of the steering vector of the transmitting array are guaranteed, and the first intermediate function A is converted from a maximum value function into a continuous function, which is convenient for subsequent calculation and processing.

[0052] Further, the step S4 specifically includes:

[0053] A second auxiliary variable h is introduced, and the second intermediate function B is transformed into a third intermediate function C using the second auxiliary variable h. The third intermediate function C is:

[0054]

[0055] in, is the maximum power data C1 in the sidelobe area of ​​the third intermediate function C, is the main lobe area minimum power data C2 of the third intermediate function C;

[0056] The second auxiliary variable h is used to transform the original constraint condition into a new constraint condition, and the new constraint condition is:

[0057] sh=0

[0058]

[0059] s H h=P

[0060] Based on the linear approximation method of the logarithmic function and the quadratic approximation method of the exponential function, the maximum power data C1 in the sidelobe area and the minimum power data C2 in the mainlobe area contained in the third intermediate function C are approximated to obtain the approximated maximum power data C1 in the sidelobe area and the approximated minimum power data C2 in the mainlobe area, wherein the approximated maximum power data C1 in the sidelobe area is:

[0061]

[0062] The main lobe area minimum power data C2 after the approximate processing is:

[0063]

[0064] in, constant is a constant, and the superscript n indicates the nth iteration;

[0065] According to the approximated sidelobe area maximum power data C1 and the approximated mainlobe area minimum power data C2, a new objective function G is obtained through the secondary optimization model. The new objective function G is:

[0066]

[0067] According to the new objective function G and the new constraint conditions, a new optimization model is constructed, and the new optimization model is:

[0068]

[0069] The beneficial effects of adopting the above further scheme are: by introducing the second auxiliary variable and adding the equality constraint sh=0, the consistency of the second auxiliary variable and the initial transmission signal matrix vector quantizer is guaranteed; by introducing the second auxiliary variable, the second intermediate function B is converted from a fourth-order function to a second-order function, which ensures the existence of a unique minimum value of the third intermediate function C; by approximating the maximum power data C1 in the sidelobe area and the minimum power data C2 in the mainlobe area contained in the third intermediate function C from a logarithmic function to an exponential function, it is ensured that the new objective function G obtained has continuity, and the new objective function G is an equation, which is convenient for the subsequent use of ADMM to solve the new optimization model.

[0070] In order to solve the above technical problems, the present invention also provides a direct waveform design device for a MIMO transmission pattern, the device comprising:

[0071] A data acquisition module, used to obtain a precoding matrix of a transmitted signal and a steering vector of a transmit array, obtain an initial transmit signal matrix vectorization operator and a covariance matrix of the steering vector of the transmit array according to the precoding matrix of the transmitted signal and the steering vector of the transmit array, and obtain initial sidelobe area power data and initial mainlobe area power data according to the initial transmit signal matrix vectorization operator and the covariance matrix;

[0072] A model initialization module, used to construct an initial optimization model according to the initial transmission signal matrix vectorization operator, the initial sidelobe area power data and the initial mainlobe area power data, wherein the initial optimization model includes an original objective function and an original constraint condition, wherein the original objective function is used to characterize the relationship between the initial sidelobe area power data and the initial mainlobe area power data, and the original constraint condition includes a sidelobe constraint, a mainlobe constraint and a total power constraint, wherein the sidelobe constraint is used to limit the power sum of the sidelobe area, the mainlobe constraint is used to limit the power sum of the mainlobe area, and the total power constraint is used to limit the transmission power of the transmitting antenna;

[0073] A model optimization module is used to construct a first auxiliary variable, use the first auxiliary variable to simplify the original objective function to obtain a first intermediate function A, the first intermediate function A includes the maximum power data A in the sidelobe area and the minimum power data A in the mainlobe area, optimize the maximum power data A in the sidelobe area and the minimum power data A in the mainlobe area contained in the first intermediate function A to obtain a second intermediate function B, the second intermediate function B includes the maximum power data B in the sidelobe area and the minimum power data B in the mainlobe area, and construct a secondary optimization model according to the second intermediate function B and the original constraint conditions;

[0074] A model transformation module is used to introduce a second auxiliary variable, and transform the secondary optimization model using the second auxiliary variable to obtain a third intermediate function C and new constraints, wherein the third intermediate function C includes the maximum power data C1 in the sidelobe area and the minimum power data C2 in the mainlobe area; the maximum power data C1 in the sidelobe area and the minimum power data C2 in the mainlobe area included in the third intermediate function C are approximated to obtain a new objective function G, and a new optimization model is constructed according to the new objective function G and the new constraints;

[0075] The model solving module is used to solve the new optimization model using the alternating direction multiplication method to obtain a target transmission signal matrix vectorization operator, calculate the target main lobe area power and the target side lobe area power according to the target transmission signal matrix vectorization operator, and draw a target transmission direction diagram according to the target main lobe area power and the target side lobe area power.

[0076] To solve the above technical problems, the present invention also provides an electronic device, which includes: a memory, a processor, and a computer program stored in the memory and executable on the processor. When the processor executes the computer program, the direct waveform design method for the MIMO transmission pattern as described above is implemented.

[0077] To solve the above technical problems, the present invention also provides a computer-readable storage medium, on which a computer program is stored. When the computer program is executed by a processor, the direct waveform design method for the MIMO transmission pattern as described above is implemented. BRIEF DESCRIPTION OF THE DRAWINGS

[0078] Figure 1 is a flow chart of the method of the present invention;

[0079] Figure 2 A diagram showing the matching effect between the target emission pattern generated by the present invention in a single peak mode and the expected emission pattern;

[0080] Figure 3A diagram showing the matching effect between the target emission pattern and the expected emission pattern generated in the multi-peak mode of the present invention;

[0081] Figure 4 A matching effect diagram of a target emission pattern generated by the present invention in a single-peak mode, a target emission pattern generated by an existing method, and an expected emission pattern;

[0082] Figure 5 A matching effect diagram of a target emission pattern generated by the present invention in a multi-peak mode, a target emission pattern generated by an existing method, and an expected emission pattern;

[0083] Figure 6 A comparison diagram of a target emission pattern generated by the present invention in a single peak mode and a target emission pattern generated by an existing method;

[0084] Figure 7 A comparison diagram of a target emission pattern generated by the present invention in a multi-peak mode and a target emission pattern generated by an existing method;

[0085] Figure 8 It is a schematic diagram of the structure of the device of the present invention;

[0086] Fig. 9 It is a schematic diagram of the structure of the electronic device of the present invention. DETAILED DESCRIPTION

[0087] The principles and features of the present invention are described below. The examples given are only used to explain the present invention and are not used to limit the scope of the present invention.

[0088] Embodiment 1

[0089] This embodiment provides a direct waveform design method for a MIMO transmission pattern, such as Figure 1 As shown, including:

[0090] Step S1: Obtain the precoding matrix of the transmit signal and the steering vector of the transmit array, obtain the covariance matrix of the initial transmit signal matrix vectorizer and the steering vector of the transmit array based on the precoding matrix of the transmit signal and the steering vector of the transmit array, and obtain initial sidelobe area power data and initial mainlobe area power data based on the covariance matrix of the initial transmit signal matrix vectorizer and the steering vector of the transmit array.

[0091] Step S1 specifically includes:

[0092] The transmitting antenna of the centralized MIMO radar is composed of M uniform linear arrays (i.e., transmitting array elements), and the transmitted signal is a wide and narrow signal. When the target is located in the far field of the centralized MIMO radar (the field far away from the antenna is called the far field), the angle of the target relative to the transmitting array element is θ, then the spatial composite signal located at the transmitting array element θ is:

[0093] y T (θ) = a T (θ)S

[0094] Where S represents the precoding matrix of the transmitted signal, and the precoding matrix of the transmitted signal S = [S1, S2, ..., S i ,...,S L ],S i =[s1(l),s2(l),...,s m (l),...,s M (l)] T , where i, m, and l are variables and their values ​​are positive integers, 1≤i≤L, 1≤m≤M, 1≤l≤L, L represents the signal code length, M represents the total number of transmitting array elements, and s m (l) represents the transmission signal of the mth transmitting array element at time l, S i represents the transmission sequence of M transmitting array elements at time l; a(θ) represents the steering vector of the transmitting array, which includes the steering vector of the sidelobe area and the steering vector of the main lobe area. d represents the array element spacing, λ represents the signal wavelength, j represents a unit imaginary number, a T (θ) represents the transpose of the steering vector of the transmitting array; y(θ) represents the power vector of the transmitting signal at an angle of θ, T (θ) represents the transposition of the power vector. In the present invention, (·) T represents the transpose operation, (·) H Represents the conjugate transpose operation.

[0095] The transmission signal power p(θ) of the centralized MIMO radar in the θ direction is:

[0096] p(θ)=y H (θ)y(θ)=(S T a(θ) H (S T a(θ)

[0097] =a H (θ)SS H a(θ)

[0098] In the above formula, y H (θ) represents the conjugate transpose of the power vector, S T represents the transpose of the precoding matrix of the transmitted signal, S H represents the conjugate transpose of the precoding matrix of the transmitted signal, a H (θ) represents the conjugate transpose of the steering vector of the transmit array.

[0099] The initial transmit signal matrix vectorization operator is:

[0100] s=vec(S)

[0101] Where vec(·) represents the matrix straightening operation, and the transmitted signal power p(θ) is:

[0102]

[0103] Among them, I L is the unit matrix of dimension L; R(θ)=a(θ)a H (θ) represents the covariance matrix of the steering vector of the transmit array.

[0104] According to the initial transmit signal matrix vector quantization operator and the covariance matrix of the steering vector of the transmit array, the initial sidelobe area power data is obtained. The initial sidelobe area power data is Among them, Θ s Represents the side lobe area, and the power in the angle area corresponding to the side lobe area is θ s represents the sidelobe area angle variable and satisfies θ s ∈Θ s , Represents the covariance matrix of the steering vectors in the sidelobe region.

[0105] According to the initial transmit signal matrix vectorization operator and the covariance matrix of the steering vector of the transmit array, the initial main lobe area power data is obtained. The initial main lobe area power data is Among them, Θ z Represents the main lobe area, and the power in the angle area corresponding to the main lobe area is θ z represents the main lobe area angle variable, and satisfies θ s ∈Θ s , Represents the covariance matrix of the steering vectors in the mainlobe region.

[0106] Step S2: construct an initial optimization model based on the initial sidelobe area power data and the initial mainlobe area power data, the initial optimization model includes the original objective function and the original constraint conditions, the original objective function is used to characterize the relationship between the initial sidelobe area power data and the initial mainlobe area power data, the original constraint conditions include sidelobe constraints, main lobe constraints and total power constraints, the sidelobe constraints are used to limit the power sum in the sidelobe area, the main lobe constraints are used to limit the power sum in the mainlobe area, and the total power constraints are used to limit the transmission power of the transmitting antenna.

[0107] Step S2 specifically includes:

[0108] The original objective function is established based on the minimum ratio of the maximum power in the side lobe area to the minimum power in the main lobe area. The original objective function is:

[0109]

[0110] According to the power in the angle area corresponding to the sidelobe area, the sidelobe constraint is established. The sidelobe constraint is:

[0111]

[0112] According to the power in the angle area corresponding to the main lobe area, the main lobe constraint is established. The main lobe constraint is:

[0113]

[0114] According to the initial transmission signal matrix vectorization operator, the total power constraint is established, and the total power constraint is:

[0115] s2=P

[0116] Among them, σ1 represents the sum of the powers in the sidelobe area, σ2 represents the sum of the powers in the mainlobe area, and P represents the total power of the transmitting array element.

[0117] According to the original objective function, side lobe constraints, main lobe constraints, and total power constraints, an initial optimization model is constructed. The initial optimization model is:

[0118]

[0119] Step S3: construct a first auxiliary variable, use the first auxiliary variable to simplify the original objective function, and obtain a first intermediate function A, the first intermediate function A includes the maximum power data A in the sidelobe area and the minimum power data A in the mainlobe area, optimize the maximum power data A in the sidelobe area and the minimum power data A in the mainlobe area contained in the first intermediate function A, and obtain a second intermediate function B, the second intermediate function B includes the maximum power data B in the sidelobe area and the minimum power data B in the mainlobe area, and construct a secondary optimization model according to the second intermediate function B and the original constraints.

[0120] Step S3 specifically includes:

[0121] Construct the first auxiliary variable g, and transform the initial sidelobe area power data and the initial mainlobe area power data in the original objective function through the first auxiliary variable g. Therefore, based on the Dinkelbach algorithm, the original objective function can be transformed to obtain the transformation function T1, which is:

[0122]

[0123] Where g is a positive constant.

[0124] The transformation function T is equivalently transformed to obtain the first intermediate function A, which is:

[0125]

[0126] in, is the maximum power data A in the sidelobe area in the first intermediate function A, is the minimum power data A in the main lobe area in the first intermediate function A.

[0127] Based on the LES method, the maximum power data A in the sidelobe area and the minimum power data A in the mainlobe area in the first intermediate function A are transformed to obtain a transformation function T2, which is:

[0128]

[0129] Here, α is a positive constant.

[0130] Using the exponential logarithmic function operation, the transformation function T2 is transformed:

[0131]

[0132] in, α and ε are both positive constants, and I represents the identity matrix;

[0133] The second intermediate function B is:

[0134]

[0135] in, is the maximum power data B in the sidelobe area in the second intermediate function B, is the minimum power data B in the main lobe area in the second intermediate function B.

[0136] According to the second intermediate function B and the original constraints, a secondary optimization model is constructed. The secondary optimization model is:

[0137]

[0138] Step S4: Introduce a second auxiliary variable, and use the second auxiliary variable to transform the quadratic optimization model to obtain a third intermediate function C and new constraints, where the third intermediate function C includes the maximum power data C1 in the sidelobe area and the minimum power data C2 in the mainlobe area; approximate the maximum power data C1 in the sidelobe area and the minimum power data C2 in the mainlobe area contained in the third intermediate function C to obtain a new objective function G, and construct a new optimization model based on the new objective function G and the new constraints.

[0139] Step S4 specifically includes:

[0140] The second auxiliary variable h is introduced, and the second auxiliary variable h is used to transform the quadratic optimization model to obtain the third intermediate function C and new constraints. The third intermediate function C is:

[0141]

[0142] in, is the maximum power data C1 of the sidelobe area in the third intermediate function C, It is the minimum power data C2 of the main lobe area in the third intermediate function C.

[0143] The new constraints are:

[0144] sh=0

[0145]

[0146] s H h=P

[0147] The maximum power data C1 of the sidelobe area after approximate processing is:

[0148]

[0149] The minimum power data C2 in the main lobe area after approximate processing is:

[0150]

[0151] in, constant is a constant, and the superscript n indicates the nth iteration;

[0152] Substitute the approximate processed sidelobe area maximum power data C1 and the approximate processed mainlobe area minimum power data C2 into the quadratic optimization model, ignore the constants, and obtain the new objective function G. The new objective function G is:

[0153]

[0154] According to the new objective function G and new constraints, a new optimization model is constructed. The new optimization model is:

[0155]

[0156] Step S5: Use ADMM to solve the new optimization model to obtain the target transmit signal matrix vectorization operator, calculate the target main lobe area power and the target side lobe area power according to the target transmit signal matrix vectorization operator, and draw the target transmit radiation pattern according to the target main lobe area power and the target side lobe area power.

[0157] Step S5 specifically includes:

[0158] make:

[0159]

[0160] The Lagrangian function is constructed according to the new optimization model. The Lagrangian function is:

[0161]

[0162] In the above formula, ρ1, ρ2, ρ3, ρ4 represent the penalty factors of each constraint condition, and u, τ1, τ2, v represent the Lagrange multipliers;

[0163] Initialization parameters 0 、h 0 , u 0 、v 0 , σ1, σ2, ρ1, ρ2, ρ3, ρ4, T, and assign initial values ​​to each parameter, where T is the total number of iterations; let the independent variable n = 0;

[0164]

[0165] u (n+1) =u n +s (n+1) -h (n+1)

[0166]

[0167] v (n+1) =v n +(s (n+1) ) H h (n+1) -P

[0168] Wherein, the superscript n indicates the nth iteration;

[0169] Iterate T rounds and update h in sequence (n+1) 、s (n+1) 、u (n+1) , v (n+1) , each update round makes n = n + 1, until n = T, the output s = s (n+1) The output s is the target transmission signal matrix vectorization operator.

[0170] For h (n+1) The calculation is done by solving We can get:

[0171] h (n+1 )=Ξ -1 γ

[0172] in:

[0173]

[0174] The computational complexity of this method increases linearly with the number of iterations: for each update h, we first need to calculate the parameter a z n 、a s n , a z n 、a s n The computational complexity is Ο(KM 2 L 2 ), then we need to calculate the matrix Ξ, the computational complexity of Ξ is Ο(KM 2 L 2 ), where K is the number of discrete sampling points about the angle, and then γ is calculated. The computational complexity of γ is O(M 2 L 2 ), and finally we need to calculate the inverse of the matrix Ξ, which has a computational complexity of Ο(M 3 L 3 ), the total complexity of updating h is O(2KM 2 L 2 +M 2 L 2 +M 3 L 3 ); The complexity of updating s is the same as h, so the total complexity of solving the new optimization model by this method is Ο(2(2KM 2 L 2 +M 2 L 2 +M 3 L 3 )). The computational complexity of using the interior point method to solve the new optimization model is O(((ML) 2 ) 3.5 ), by comparison, the complexity of this method is lower.

[0175] Simulation experiment

[0176] The performance of the method of this embodiment is evaluated by computer simulation, as follows:

[0177] Assume that both the MIMO radar transmitting array and receiving array use uniform linear arrays, the array element spacing is half a wavelength, the number of transmitting array elements M = 10, the signal code length of each transmitting array element L = 32, the carrier frequency f0 = 3GHz, the expected main lobe amplitude normalized gain is 1, and the side lobe area gain is 0. Consider two different pattern matching modes: (1) Single peak mode: Θ z =[-30°,30°], the number of discretization points is 61, and the sidelobe area Θ s=[-90°,-35°]∪[35°,90°], the number of discretization points is 112; (2) Multi-peak mode: main lobe area Θ z =[-70°,-50°]∪[50°,70°], the number of discretization points is 41, and the sidelobe area Θ s =[-90°,-75°]∪[-45°,45°]∪[75°,90°], the number of discretization points is 123.

[0178] The centralized MIMO radar target emission pattern generated by this method is as follows: Figure 2 and Figure 3 As shown, according to Figure 2 It can be seen that in the single-peak mode, the maximum power of the main lobe area of ​​the target transmission pattern obtained by this method is 1.03, the minimum power of the main lobe area is 0.98, the fluctuation error is less than 5%, the main lobe fluctuation is small, and the side lobe power is compressed below 0.05. The power data of the main lobe area is much larger than the power data of the side lobe area. Compared with the main lobe area power fixed value of 1 and the side lobe area power fixed value of 0 of the expected transmission pattern, this method can effectively improve the main lobe gain and reduce the side lobe peak value, and the side lobe suppression effect is obvious; according to Figure 3 It can be seen that in the multi-peak mode, the maximum power of the main lobe area of ​​the target transmission pattern obtained by this method is about 4.4, and it is a single peak, the power of the side area is below 0.1, and the power data of the main lobe area is much larger than the power data of the side lobe area. Compared with the main lobe area power fixed value of 4 and the side lobe area power fixed value of 0.1 of the expected transmission pattern, this method can effectively improve the main lobe gain and reduce the side lobe peak value, and the side lobe suppression effect is obvious; according to Figure 2 and Figure 3 It can be seen that in both single-peak mode and multi-peak mode, the target emission pattern generated by this method matches the expected emission pattern well. This method can effectively improve the main lobe gain and reduce the peak side lobe, and impose certain restrictions on the power of the two, and has a significant effect on suppressing the side lobe.

[0179] The matching effect diagram of the centralized MIMO radar target emission pattern generated by this method, the target emission pattern generated by the method of minimizing the template matching error and the expected emission pattern is shown in the figure below. Figure 4 and Figure 5 As shown, Figure 4 and Figure 5 In the example, target emission pattern A is the target emission pattern generated by this method, and target emission pattern B is the target emission pattern generated by minimizing the template matching error method. Figure 4 and Figure 5It can be seen that both methods can achieve relatively good transmission pattern matching effects in single-peak mode and double-peak mode, but compared with the method of minimizing template matching error, this method has lower main lobe fluctuation. After normalizing the power and taking the logarithm, as shown in Figure 6 and Figure 7 As shown, Figure 6 and Figure 7 In the example, target emission pattern A is the target emission pattern generated by this method, and target emission pattern B is the target emission pattern generated by minimizing the template matching error method. Figure 6 and Figure 7 It can be seen that in single-peak mode, the peak sidelobe power of this method is -25.62dB, while the peak sidelobe power of the method of minimizing the template matching error is -13.20dB. This method can reduce the sidelobe power by about 12dB; in multi-peak mode, the peak sidelobe power of this method is -16.5dB, while the peak sidelobe power of the method of minimizing the template matching error is -8dB. This method can reduce the sidelobe power by about 8dB. It can be seen that this method can better reduce the sidelobe power.

[0180] Embodiment 2

[0181] This embodiment provides a direct waveform design device for a MIMO transmission pattern, such as Figure 8 As shown, the device comprises:

[0182] A data acquisition module, used to obtain a precoding matrix of a transmitted signal and a steering vector of a transmitting array, obtain an initial transmit signal matrix vectorization operator and a covariance matrix of the steering vector of the transmitting array according to the precoding matrix of the transmitted signal and the steering vector of the transmitting array, and obtain initial sidelobe area power data and initial mainlobe area power data according to the covariance matrix of the initial transmit signal matrix vectorization operator and the steering vector of the transmitting array;

[0183] A model initialization module is used to construct an initial optimization model according to an initial transmission signal matrix vectorization operator, initial sidelobe area power data and initial mainlobe area power data. The initial optimization model includes an original objective function and original constraints. The original objective function is used to characterize the relationship between the initial sidelobe area power data and the initial mainlobe area power data. The original constraints include sidelobe constraints, mainlobe constraints and total power constraints. The sidelobe constraints are used to limit the power sum in the sidelobe area, the mainlobe constraints are used to limit the power sum in the mainlobe area, and the total power constraints are used to limit the transmission power of the transmitting antenna.

[0184] A model optimization module is used to construct a first auxiliary variable, simplify the original objective function using the first auxiliary variable to obtain a first intermediate function A, the first intermediate function A includes the maximum power data A in the sidelobe area and the minimum power data A in the mainlobe area, optimize the maximum power data A in the sidelobe area and the minimum power data A in the mainlobe area contained in the first intermediate function A to obtain a second intermediate function B, the second intermediate function B includes the maximum power data B in the sidelobe area and the minimum power data B in the mainlobe area, and construct a secondary optimization model according to the second intermediate function B and the original constraint conditions;

[0185] A model transformation module is used to introduce a second auxiliary variable, transform the secondary optimization model using the second auxiliary variable, obtain a third intermediate function C and new constraints, the third intermediate function C includes the maximum power data C1 in the sidelobe area and the minimum power data C2 in the mainlobe area; perform approximate processing on the maximum power data C1 in the sidelobe area and the minimum power data C2 in the mainlobe area included in the third intermediate function C to obtain a new objective function G, and construct a new optimization model according to the new objective function G and the new constraints;

[0186] The model solving module is used to solve the new optimization model by using the alternating direction multiplication method to obtain the target transmission signal matrix vectorization operator, calculate the target main lobe area power and the target side lobe area power according to the target transmission signal matrix vectorization operator, and draw the target transmission direction diagram according to the target main lobe area power and the target side lobe area power.

[0187] The data acquisition module includes a first data acquisition unit, a second data acquisition unit and a third data acquisition unit, wherein:

[0188] The first data acquisition unit is used to acquire the precoding matrix of the transmission signal of the centralized MIMO radar transmission array composed of multiple transmission array elements in the target direction θ and the steering vector of the transmission array;

[0189] The second data acquisition unit is used to acquire the covariance matrix of the initial transmission signal matrix vector quantizer and the steering vector of the transmission array according to the precoding matrix of the transmission signal and the steering vector of the transmission array;

[0190] The third data acquisition unit is used to acquire initial sidelobe area power data and initial mainlobe area power data according to the initial transmit signal matrix quantization operator and the covariance matrix of the steering vector of the transmit array.

[0191] The model initialization module includes an objective function building unit, a constraint building unit and a model building unit A, where:

[0192] The objective function building unit is used to establish the original objective function according to the minimum ratio of the maximum power in the side lobe area to the minimum power in the main lobe area;

[0193] The constraint construction unit is used to establish the original constraint conditions according to the initial transmission signal matrix vectorization operator, the initial side lobe area power data and the initial main lobe area power data: establish the side lobe constraint according to the power of the angle area corresponding to the side lobe area, establish the main lobe constraint according to the power of the angle area corresponding to the main lobe area, and establish the total power constraint according to the initial transmission signal matrix vectorization operator;

[0194] The model building unit A is used to build an initial optimization model according to the original objective function and the original constraint conditions.

[0195] The model optimization module includes an objective function simplification unit, an objective function optimization unit and a model building unit B, wherein:

[0196] The objective function simplification unit is used to construct a first auxiliary variable, and the initial sidelobe area power data and the initial mainlobe area power data in the original objective function are simplified by the first auxiliary variable to obtain a first intermediate function A, wherein the first intermediate function A includes the maximum power data A in the sidelobe area and the minimum power data A in the mainlobe area;

[0197] The objective function optimization unit is used to optimize the maximum power data A in the sidelobe area and the minimum power data A in the mainlobe area in the first intermediate function A based on the LES method and the exponential logarithmic function operation to obtain the second intermediate function B, wherein the second intermediate function B includes the maximum power data B in the sidelobe area and the minimum power data B in the mainlobe area;

[0198] The model building unit B is used to build a secondary optimization model according to the second intermediate function B and the original constraint conditions.

[0199] The model transformation module includes a replacement unit, an objective function transformation unit and a model construction unit C, wherein:

[0200] The replacement unit is used to introduce a second auxiliary variable, and transform the secondary optimization model using the second auxiliary variable to obtain a third intermediate function C and a new constraint condition, wherein the third intermediate function C includes maximum power data C1 in the side lobe area and minimum power data C2 in the main lobe area;

[0201] The objective function conversion unit is used to perform approximate processing on the maximum power data C1 in the side lobe area and the minimum power data C2 in the main lobe area included in the third intermediate function C to obtain a new objective function G;

[0202] The model building unit C is used to build a new optimization model according to the new objective function G and new constraints.

[0203] The model solving module includes a model solving unit, a power solving unit and a directional pattern forming unit, wherein:

[0204] The model solving unit is used to introduce Lagrangian multipliers and penalty factors, construct Lagrangian functions according to the new optimization model, solve the Lagrangian functions using the alternating direction multiplier method, and obtain the target transmission signal matrix vectorization operator;

[0205] The power solving unit is used to calculate the target main lobe area power and the target side lobe area power according to the target transmission signal matrix vector quantization operator;

[0206] The pattern forming unit is used to draw the target transmission pattern according to the target main lobe area power and the target side lobe area power.

[0207] Embodiment 3

[0208] This embodiment provides an electronic device, such as Fig. 9 As shown, the electronic device includes: a memory, a processor, and a computer program stored in the memory and executable on the processor. When the processor executes the computer program, the direct waveform design method for the MIMO transmission pattern as described in the first embodiment is implemented.

[0209] Embodiment 4

[0210] This embodiment provides a computer-readable storage medium, on which a computer program is stored. When the computer program is executed by a processor, the direct waveform design method for the MIMO transmission pattern as described in the first embodiment is implemented.

[0211] Although the embodiments of the present invention have been shown and described above, it is to be understood that the above embodiments are exemplary and are not to be construed as limitations of the present invention. A person skilled in the art may change, modify, replace and vary the above embodiments within the scope of the present invention.

Claims

1. A direct waveform design method for MIMO transmission pattern, characterized in that: include: Step S1: obtaining a precoding matrix of a transmitted signal and a steering vector of a transmitting array, obtaining an initial transmitted signal matrix vectorization operator and a covariance matrix of the steering vector of the transmitting array according to the precoding matrix of the transmitted signal and the steering vector of the transmitting array, and obtaining initial sidelobe area power data and initial mainlobe area power data according to the initial transmitted signal matrix vectorization operator and the covariance matrix; Step S2: constructing an initial optimization model according to the initial transmission signal matrix vectorization operator, the initial sidelobe area power data and the initial mainlobe area power data, wherein the initial optimization model includes an original objective function and an original constraint condition, wherein the original objective function is used to characterize the relationship between the initial sidelobe area power data and the initial mainlobe area power data, and the original constraint condition includes a sidelobe constraint, a mainlobe constraint and a total power constraint, wherein the sidelobe constraint is used to limit the power sum in the sidelobe area, the mainlobe constraint is used to limit the power sum in the mainlobe area, and the total power constraint is used to limit the transmission power of the transmitting antenna; Step S3: construct a first auxiliary variable, use the first auxiliary variable to simplify the original objective function to obtain a first intermediate function A, the first intermediate function A includes the maximum power data A in the sidelobe area and the minimum power data A in the mainlobe area, optimize the maximum power data A in the sidelobe area and the minimum power data A in the mainlobe area contained in the first intermediate function A to obtain a second intermediate function B, the second intermediate function B includes the maximum power data B in the sidelobe area and the minimum power data B in the mainlobe area, and construct a secondary optimization model according to the second intermediate function B and the original constraint conditions; Step S4: introducing a second auxiliary variable, using the second auxiliary variable to transform the secondary optimization model to obtain a third intermediate function C and new constraints, wherein the third intermediate function C includes the maximum power data C1 in the sidelobe area and the minimum power data C2 in the mainlobe area; performing approximate processing on the maximum power data C1 in the sidelobe area and the minimum power data C2 in the mainlobe area contained in the third intermediate function C to obtain a new objective function G, and constructing a new optimization model according to the new objective function G and the new constraints; Step S5: Use the alternating direction multiplication method to solve the new optimization model to obtain the target transmission signal matrix vectorization operator, calculate the target main lobe area power and the target side lobe area power according to the target transmission signal matrix vectorization operator, and draw the target transmission radiation pattern according to the target main lobe area power and the target side lobe area power.

2. A direct waveform design method for a MIMO transmission pattern according to claim 1, characterized in that: The step S1 specifically includes: The precoding matrix of the transmission signal of the centralized MIMO radar transmission array composed of multiple transmission array elements in the target direction θ and the steering vector of the transmission array are obtained, wherein the precoding matrix of the transmission signal is: S=[S1,S2,...,S i ,...,S L ] Wherein, S represents the precoding matrix of the transmitted signal, S i =[s1(l),s2(l),...,s m (l),...,s M (l)] T , i, m, l are all variables and their values ​​are positive integers, 1≤i≤L, 1≤m≤M, 1≤l≤L, L represents the signal code length, M represents the total number of transmitting array elements, s m (l) represents the transmission signal of the mth transmitting array element at time l, S i represents the transmission sequence of multiple transmitting array elements at time l; The steering vector of the transmitting array is: Wherein, a(θ) represents the steering vector of the transmitting array, the steering vector of the transmitting array includes the steering vector of the side lobe area and the steering vector of the main lobe area, d represents the array element spacing, λ represents the signal wavelength, and j represents a unit imaginary number; (·) T Represents the transpose operation; An initial transmit signal matrix vectorization operator is obtained according to the precoding matrix of the transmit signal, and the initial transmit signal matrix vectorization operator is: s=vec(S) Wherein, s represents the matrix vectorization operator of the initial transmission signal, and vec(·) represents the matrix straightening operation; The covariance matrix of the steering vector of the transmit array is obtained according to the steering vector of the transmit array. The covariance matrix of the steering vector of the transmit array is: R(θ)=a(θ)a H (i) Where R(θ) represents the covariance matrix of the steering vector of the transmit array, (·) H represents the conjugate transpose operation; According to the initial transmission signal matrix vectorization operator and the covariance matrix, the initial sidelobe area power data is obtained, and the initial sidelobe area power data is Among them, Θ s Represents the side lobe area, and the power in the angle area corresponding to the side lobe area is θ s represents the sidelobe area angle variable and satisfies θ s ∈Θ s , represents the covariance matrix of the steering vectors in the sidelobe region; According to the initial transmission signal matrix vectorization operator and the covariance matrix, the initial main lobe area power data is obtained, and the initial main lobe area power data is Among them, Θ z Represents the main lobe area, and the power in the angle area corresponding to the main lobe area is θ z represents the main lobe area angle variable, and satisfies θ s ∈Θ s , Represents the covariance matrix of the steering vectors in the mainlobe region.

3. The direct waveform design method of a MIMO transmission pattern according to claim 2, characterized in that: In the step S2, constructing the initial optimization model according to the initial transmit signal matrix vector quantizer, the initial sidelobe area power data and the initial mainlobe area power data specifically includes: The original objective function is established according to the minimum ratio of the maximum power in the side lobe area to the minimum power in the main lobe area, and the original objective function is: A sidelobe constraint is established according to the power in the angle region corresponding to the sidelobe region, and the sidelobe constraint is: According to the power in the angle region corresponding to the main lobe region, a main lobe constraint is established, and the main lobe constraint is: According to the initial transmission signal matrix vectorization operator, a total power constraint is established, and the total power constraint is: s2=P Where σ1 represents the sum of the power in the sidelobe area, σ2 represents the sum of the power in the mainlobe area, and P represents the total power of the transmitting array element; According to the original objective function, the side lobe constraint, the main lobe constraint, and the total power constraint, the initial optimization model is constructed, and the initial optimization model is:

4. The direct waveform design method of a MIMO transmission pattern according to claim 3, characterized in that: The step S3 specifically includes: The first auxiliary variable g is constructed, and the initial sidelobe area power data and the initial mainlobe area power data in the original objective function are transformed by the first auxiliary variable g to obtain the first intermediate function A, where the first intermediate function A is: in, is the maximum power data A in the sidelobe area in the first intermediate function A, is the minimum power data A in the main lobe area in the first intermediate function A; Based on the LES method and the exponential logarithmic function operation, the maximum power data A in the sidelobe area and the minimum power data A in the mainlobe area in the first intermediate function A are optimized to obtain the second intermediate function B, and the second intermediate function B is: in, α and ε are both positive constants, and I represents the identity matrix; is the maximum power data B in the sidelobe area in the second intermediate function B, is the minimum power data B in the main lobe area in the second intermediate function B; According to the second intermediate function B and the original constraint condition, the secondary optimization model is constructed, and the secondary optimization model is:

5. The direct waveform design method of a MIMO transmission pattern according to claim 4, characterized in that: The step S4 specifically includes: A second auxiliary variable h is introduced, and the second intermediate function B is transformed into a third intermediate function C using the second auxiliary variable h. The third intermediate function C is: in, is the maximum power data C1 in the sidelobe area of ​​the third intermediate function C, is the main lobe area minimum power data C2 of the third intermediate function C; The second auxiliary variable h is used to transform the original constraint condition into a new constraint condition, and the new constraint condition is: sh=0 s H h=P Based on the linear approximation method of the logarithmic function and the quadratic approximation method of the exponential function, the maximum power data C1 in the sidelobe area and the minimum power data C2 in the mainlobe area contained in the third intermediate function C are approximated to obtain the approximated maximum power data C1 in the sidelobe area and the approximated minimum power data C2 in the mainlobe area, wherein the approximated maximum power data C1 in the sidelobe area is: The main lobe area minimum power data C2 after the approximate processing is: in, constant is a constant, and the superscript n indicates the nth iteration; According to the approximated sidelobe area maximum power data C1 and the approximated mainlobe area minimum power data C2, a new objective function G is obtained through the secondary optimization model. The new objective function G is: According to the new objective function G and the new constraint conditions, a new optimization model is constructed, and the new optimization model is:

6. A direct waveform design device for MIMO transmission pattern, characterized in that: include: A data acquisition module, used to obtain a precoding matrix of a transmitted signal and a steering vector of a transmit array, obtain an initial transmit signal matrix vectorization operator and a covariance matrix of the steering vector of the transmit array according to the precoding matrix of the transmitted signal and the steering vector of the transmit array, and obtain initial sidelobe area power data and initial mainlobe area power data according to the initial transmit signal matrix vectorization operator and the covariance matrix; A model initialization module, used to construct an initial optimization model according to the initial transmission signal matrix vectorization operator, the initial sidelobe area power data and the initial mainlobe area power data, wherein the initial optimization model includes an original objective function and an original constraint condition, wherein the original objective function is used to characterize the relationship between the initial sidelobe area power data and the initial mainlobe area power data, and the original constraint condition includes a sidelobe constraint, a mainlobe constraint and a total power constraint, wherein the sidelobe constraint is used to limit the power sum of the sidelobe area, the mainlobe constraint is used to limit the power sum of the mainlobe area, and the total power constraint is used to limit the transmission power of the transmitting antenna; A model optimization module is used to construct a first auxiliary variable, use the first auxiliary variable to simplify the original objective function to obtain a first intermediate function A, the first intermediate function A includes the maximum power data A in the sidelobe area and the minimum power data A in the mainlobe area, optimize the maximum power data A in the sidelobe area and the minimum power data A in the mainlobe area contained in the first intermediate function A to obtain a second intermediate function B, the second intermediate function B includes the maximum power data B in the sidelobe area and the minimum power data B in the mainlobe area, and construct a secondary optimization model according to the second intermediate function B and the original constraint conditions; A model transformation module is used to introduce a second auxiliary variable, and transform the secondary optimization model using the second auxiliary variable to obtain a third intermediate function C and new constraints, wherein the third intermediate function C includes the maximum power data C1 in the sidelobe area and the minimum power data C2 in the mainlobe area; the maximum power data C1 in the sidelobe area and the minimum power data C2 in the mainlobe area included in the third intermediate function C are approximated to obtain a new objective function G, and a new optimization model is constructed according to the new objective function G and the new constraints; The model solving module is used to solve the new optimization model using the alternating direction multiplication method to obtain a target transmission signal matrix vectorization operator, calculate the target main lobe area power and the target side lobe area power according to the target transmission signal matrix vectorization operator, and draw a target transmission direction diagram according to the target main lobe area power and the target side lobe area power.

7. An electronic device, characterized in that: include: A memory, a processor, and a computer program stored in the memory and executable on the processor, wherein when the processor executes the computer program, the direct waveform design method for the MIMO transmission pattern as claimed in any one of claims 1 to 5 is implemented.

8. A computer-readable storage medium, characterized in that: A computer program is stored on a computer-readable storage medium, and when the computer program is executed by a processor, the direct waveform design method for a MIMO transmission pattern as claimed in any one of claims 1 to 5 is implemented.

Images