Target detection method, system and device of MIMO radar and medium

Abstract

The invention relates to the technical field of radar detection, in particular to a target detection method, system and device of an MIMO radar and a medium. The method comprises the following steps: acquiring a transmitted waveform set composed of transmitted waveforms of a plurality of transmitting antennas, a mismatched filtering vector set composed of received vectors of a plurality of mismatched filters and main lobe loss in the MIMO radar; under the constraint of the amplitude of the transmitted waveform, the receiving vector of the mismatched filter and the main lobe loss, establishing a total optimization problem taking the minimum ISL value of the radar signal GCAF as a target; an MBI algorithm is adopted to decompose the total optimization problem into a first sub-optimization problem and a second sub-optimization problem; an SCA algorithm is adopted to convert the first sub-optimization problem and the second sub-optimization problem into a first convex optimization problem and a second convex optimization problem; and solving the first convex optimization problem and the second convex optimization problem to obtain the target amplitude of the transmitted waveform in the radar and the target receiving vector of the mismatched filter so as to improve the radar detection precision.

Description

Technical Field

[0001] The present invention relates to the technical field of radar signal processing, and particularly to a method, system, device and medium for target detection of a MIMO radar. Background Art

[0002] Waveform design has important application values in many fields such as radar and wireless communication. In traditional phased array radars, multiple antennas can only transmit the same waveform, while as a new type of radar system, Multiple Input Multiple Output (MIMO) radars allow different antennas to independently transmit different waveforms. If the transmitted waveforms satisfy orthogonality, the system can obtain a larger virtual aperture, thereby improving the detection performance, parameter identification performance, resolution, etc. of the system. Therefore, the orthogonal waveform design of MIMO radars has received extensive attention in recent years.

[0003] The main purpose of MIMO radar orthogonal waveform design is to suppress the waveform autocorrelation sidelobe level and cross-correlation level. The lower the autocorrelation sidelobe level, that is, the less correlated the waveform is with the waveform after its own time shift, the better the target resolution ability and synchronization performance of the system; a lower cross-correlation level means less correlation between waveforms, which can improve the parameter estimation performance and suppress system mutual interference, etc. At the same time, good auto / cross-correlation can also ensure that the radar receiving filter can more easily separate the transmission degrees of freedom from the echo signal.

[0004] Among them, the commonly used optimization criteria for measuring auto / cross-correlation include Integrated Sidelobe Level (ISL), Peak Sidelobe Level (PSL), Weighted Peak Sidelobe Level (WPSL), and Weighted Integrated Sidelobe Level (WISL). Based on the above criteria, the problem of MIMO radar orthogonal waveform design can be modeled, and waveforms with good auto- and cross-correlation characteristics can be obtained by solving.

[0005] However, when modeling the above problems currently, usually only the design of the radar transmission sequence set is considered, without considering the design of the radar receiving filter, resulting in a relatively low degree of freedom and low detection accuracy of the radar finally. Summary of the Invention

[0006] The purpose of the present invention is to provide a method, system, device and medium for target detection of a MIMO radar, which can improve the accuracy of radar target detection.

[0007] To solve the above technical problems, an embodiment of the present invention provides a target detection method for a MIMO radar, including the following steps: Obtain a set of transmitted waveform sets composed of transmitted waveforms of multiple transmitting antennas, a set of mismatched filtering vector sets composed of received vectors of multiple mismatched filters, and the main lobe loss in the MIMO radar; Under the constraints of the amplitude of the transmitted waveform, the received vector of the mismatched filter, and the main lobe loss, establish a total optimization problem with the goal of minimizing the ISL value of the generalized cross ambiguity function (GCAF) of the MIMO radar signal; Use the MBI algorithm to solve the total optimization problem. In each iteration of the MBI algorithm, by fixing the variables other than the amplitude of the transmitted waveform in the total optimization problem, or by fixing the variables other than the received vector of the fixed mismatched filter in the total optimization problem, the total optimization problem is decomposed into a first sub-optimization problem with the goal of minimizing the ISL value of the GCAF of the MIMO radar signal to optimize the received vector of the mismatched filter, and a second sub-optimization problem with the goal of minimizing the ISL value of the GCAF of the MIMO radar signal to optimize the amplitude of the transmitted waveform; Use the SCA algorithm to solve the first sub-optimization problem and the second sub-optimization problem respectively. In each iteration of the SCA algorithm, by using a preset convex approximation function to replace the non-convex objective function and non-convex constraints in the first sub-optimization problem and the second sub-optimization problem, the first sub-optimization problem and the second sub-optimization problem are respectively transformed into a first convex optimization problem and a second convex optimization problem; Solve the first convex optimization problem and the second convex optimization problem respectively to obtain the target amplitude of the transmitted waveform and the target received vector of the mismatched filter in the MIMO radar, so that the transmitting antenna and the mismatched filter of the MIMO radar jointly perform target detection with the target amplitude and the target received vector respectively.

[0008] Optionally, the GCAF of the MIMO radar signal is: ; In the formula, represents the transmitted waveform set of the th transmitting antenna, represents the mismatched filtering vector set of the th receiving antenna, represents the length of the transmitted waveform set, represents the Doppler frequency shift vector set, represents the time delay set, N -1 = K ; Simplified representation: ; In the formula, denotes the conjugate transpose operation, denotes the diagonal matrix formed by the vector ; and denote the Doppler shift vector and the matrix of time delays, respectively; ; ; The total optimization problem is expressed by the following formula: ; In the formula, C1, C2, and C3 denote the main lobe loss constraint, the amplitude constraint of the transmitted waveform, and the received vector constraint of the mismatched filter, respectively, denotes a preset main lobe gain coefficient, denotes the main lobe set, denotes the fluctuation degree of the transmitted waveform amplitude, denotes all the sidelobe sets of the local GCAF.

[0009] Optionally, the first sub-optimization problem and the second sub-optimization problem are respectively expressed by the following formulas: ; .

[0010] Optionally, the solving of the first sub-optimization problem and the second sub-optimization problem respectively includes: Obtain the first-order Taylor expansions of the three concave functions in the first sub-optimization problem and the second sub-optimization problem , and ; Substitute the first-order Taylor expansions of the three concave functions into the first sub-optimization problem and the second sub-optimization problem respectively to transform the first sub-optimization problem and the second sub-optimization problem into the first convex optimization problem and the second convex optimization problem respectively.

[0011] Optionally, the first convex optimization problem and the second convex optimization problem are respectively expressed by the following formulas: ; .

[0012] Optionally, the first convex optimization problem is solved through the following steps: For the first convex optimization problem, introduce auxiliary variables , and , and the first convex optimization problem is equivalent to: ; The augmented Lagrangian function of the first convex optimization problem after equivalence is as follows: ; In the formula, and respectively represent the Lagrange multiplier vector and the penalty factor; Based on each term in , obtain the optimization problem with respect to , the optimization problem with respect to , the optimization problem with respect to , and the optimization problem with respect to : ; In the formula, , , ; ; In the formula, , ; ; In the formula, ; ; In the formula, ; Based on the ADPM algorithm, solve the optimization problems with respect to , the optimization problems with respect to , the optimization problems with respect to , and the optimization problems with respect to respectively, so that the penalty factor in the ADPM algorithm is dynamically updated in each iteration, and the penalty term gradually approaches zero.

[0013] An embodiment of the present invention also provides a target detection system for a MIMO radar, including: A parameter acquisition module, configured to acquire a transmission waveform set composed of transmission waveforms of multiple transmit antennas in the MIMO radar, a mismatched filtering vector set composed of received vectors of multiple mismatched filters, and the main lobe loss; A problem establishment module, configured to establish a total optimization problem with the minimum ISL value of the generalized cross ambiguity function (GCAF) of the MIMO radar signal as the goal under the constraints of the amplitude of the transmission waveform, the received vector of the mismatched filter, and the main lobe loss; A problem decomposition module, which is used to solve the total optimization problem by using the MBI algorithm. In each iteration of the MBI algorithm, by fixing the variables other than the amplitude of the transmitted waveform in the total optimization problem, or fixing the variables other than the received vector of the mismatched filter in the total optimization problem, the total optimization problem is decomposed into a first sub-optimization problem that aims to minimize the ISL value of the MIMO radar signal GCAF and optimizes the received vector of the mismatched filter, and a second sub-optimization problem that aims to minimize the ISL value of the MIMO radar signal GCAF and optimizes the amplitude of the transmitted waveform; A problem transformation module, which is used to solve the first sub-optimization problem and the second sub-optimization problem respectively by using the SCA algorithm. In each iteration of the SCA algorithm, by using a preset convex approximation function to replace the non-convex objective function and non-convex constraints in the first sub-optimization problem and the second sub-optimization problem, the first sub-optimization problem and the second sub-optimization problem are respectively transformed into a first convex optimization problem and a second convex optimization problem; A problem solving module, which is used to solve the first convex optimization problem and the second convex optimization problem respectively to obtain the target amplitude of the transmitted waveform and the target received vector of the mismatched filter in the MIMO radar, so that the transmitting antenna and the mismatched filter of the MIMO radar jointly perform target detection with the target amplitude and the target received vector respectively.

[0014] An embodiment of the present invention also provides a computer device, including: at least one processor; and a memory communicatively connected to the at least one processor; wherein, the memory stores instructions executable by the at least one processor, and the instructions are executed by the at least one processor so that the at least one processor can execute the above-mentioned target detection method of the MIMO radar.

[0015] An embodiment of the present invention also provides a computer-readable storage medium storing a computer program, and when the computer program is executed by a processor, the above-mentioned target detection method of the MIMO radar is implemented.

[0016] The target detection method of the MIMO radar provided by the present invention has at least the following beneficial effects: By establishing a total optimization problem that aims to minimize the ISL value of the MIMO radar signal GCAF under the constraints of the amplitude of the MIMO radar transmitted waveform, the received vector of the mismatched filter, and the main lobe loss, and solving it, the transmitting antenna and the mismatched filter of the MIMO radar perform target detection with the solved target amplitude and target received vector respectively. This total optimization problem not only simultaneously considers the design of the radar transmission sequence set and the receiving filter, improves the degree of freedom of the radar, but also can optimize the waveform in the MIMO radar to achieve a lower ISL value and a controllable waveform peak-to-average ratio, effectively improving the accuracy of the MIMO radar target detection. Brief Description of the Drawings

[0017] One or more embodiments are exemplarily illustrated by the pictures in the corresponding drawings, and these exemplary illustrations do not constitute a limitation on the embodiments.

[0018] Figure 1 is a flowchart of a method for target detection of a MIMO radar provided according to an embodiment of the present invention; Figure 2 is a graph showing the variation of the ISL value of a different method with the number of iterations provided according to an embodiment of the present invention; Figure 3 is a three-dimensional ISL map of the range-Doppler cells obtained by a MBI-SCA method provided according to an embodiment of the present invention; Figure 4 is a three-dimensional ISL map of the range-Doppler cells obtained by a MBI-ADPM method provided according to an embodiment of the present invention; Figure 5 is a zero-Doppler two-dimensional map of a different method provided according to an embodiment of the present invention. Detailed Embodiments

[0019] To make the objectives, technical solutions, and advantages of the embodiments of the present invention clearer, the embodiments of the present invention will be described in detail below with reference to the drawings. However, those of ordinary skill in the art can understand that in the embodiments of the present invention, many technical details are provided to help the reader better understand the present invention. However, even without these technical details and various changes and modifications based on the following embodiments, the technical solutions claimed by the present invention can still be implemented. The following division of each embodiment is for convenience of description and should not constitute any limitation on the specific implementation manner of the present invention. Each embodiment can be combined and cross-referenced with each other on the premise of not being contradictory.

[0020] An embodiment of the present invention relates to a method for target detection of a MIMO radar. The implementation details of the method for target detection of the MIMO radar in this embodiment will be specifically described below. The following content is only the implementation details provided for convenience of understanding and is not necessary for implementing this solution.

[0021] The specific process of the method for target detection of the MIMO radar in this embodiment can be as Figure 1 shown and includes: Step 101, obtain a set of transmitted waveforms formed by the transmitted waveforms of multiple transmit antennas in the MIMO radar, a set of mismatched filtering vectors formed by the received vectors of multiple mismatched filters, and the main lobe loss.

[0022] Step 102: Under the constraints of the amplitude of the transmitted waveform, the received vector of the mismatch filter, and the main lobe loss, establish a total optimization problem with the goal of minimizing the ISL value of the generalized cross ambiguity function (GCAF) of the MIMO radar signal.

[0023] Specifically, assume that the number of transmitters and receivers (i.e., transmit antennas and receive antennas) of the MIMO radar system (which can be a distributed MIMO radar or a centralized MIMO radar) is both M. Denote the transmitted waveform set of the -th transmit antenna, and denote the mismatch filtering vector set of the -th receive antenna, where denotes the length of the transmitted waveform set.

[0024] At this time, the discrete GCAF of the MIMO radar signal is defined as: ; In the formula, denotes the Doppler frequency shift vector set, denotes the time delay set, N -1 = K .

[0025] For simplicity of representation, it is expressed in matrix form as: ; In the formula, denotes the conjugate transpose operation, denotes the diagonal matrix composed of the vector , and denote the matrices of the Doppler frequency shift vector and the time delay respectively.

[0026] That is: ; .

[0027] In this embodiment, considering the joint design of the transmitted waveform set and the mismatch filter bank, that is, the non-matching filtering system. Compared with the matching filtering system, the mismatch filtering can obtain a lower sidelobe level at the cost of a slight main lobe loss. Therefore, the influence of the main lobe loss brought by the mismatch filter needs to be considered during the waveform design process, and the main lobe loss can be controlled by the following formula:

[0028] ; In the formula, is the main lobe loss coefficient, and its value range is , denotes the main lobe set.

[0029] Secondly, considering engineering implementation, it is necessary to limit the amplitude of the transmitted waveform within a certain range to control the peak-to-average ratio of the waveform. The constraint on the dynamic range of the transmitted sequence amplitude can be expressed as: ; In the formula, represents the degree of fluctuation of the waveform amplitude.

[0030] In addition, considering the limited energy of the receiving filter, a total energy constraint is imposed on it, that is .

[0031] Finally, due to the fact that the Doppler shift of the target is much lower than the signal bandwidth and the equal-volume property of the Generalized Cross Ambiguity Function (GCAF), it is impossible to reduce the level of the entire range-Doppler cell simultaneously, and only the local range-Doppler region of interest can be optimized , which is defined as , , .

[0032] Based on the above, in the context of waveform design in the fast-time dimension, this embodiment adopts the ISL minimization criterion to jointly design the transmitted sequence set and the mismatched filter bank, and the total optimization problem can be modeled as: ; In the formula, represents all the sidelobe sets of the local GCAF, and .

[0033] Step 103, use the MBI algorithm to solve the total optimization problem. In each iteration of the MBI algorithm, by fixing the variables in the total optimization problem except the amplitude of the transmitted waveform, or fixing the variables in the total optimization problem except the receiving vector of the fixed mismatched filter, the total optimization problem is decomposed into a first sub-optimization problem that optimizes the receiving vector of the mismatched filter with the goal of minimizing the ISL value of the GCAF of the MIMO radar signal, and a second sub-optimization problem that optimizes the amplitude of the transmitted waveform with the goal of minimizing the ISL value of the GCAF of the MIMO radar signal.

[0034] Specifically, the objective function of the total optimization problem is a non-convex fourth-order polynomial function, and both C1 and C2 in it are non-convex constraints. Therefore, the total optimization problem is a typical HOP non-convex optimization problem. It is relatively difficult to directly solve the HOP non-convex optimization problem. However, if or is fixed, the objective function is with respect to and The variables are all quadratic convex functions. Based on this, this embodiment will solve them based on the MBI algorithm framework. The MBI algorithm can well solve the HOP optimization problem, especially suitable for scenarios where the problem variables can be naturally divided into multiple blocks. This method fixes other variable blocks in each iteration, only optimizes a certain variable block, and selects the block that improves the objective function the most for update, thereby gradually improving the quality of the solution.

[0035] Based on this idea, fix except for or of the other variables, representing the th iteration of MBI, solve the corresponding optimization problem. The optimization problems related to and can be respectively expressed as: ; .

[0036] For convenience of representation, use to represent the objective function. After obtaining the solutions and of the above two formulas, calculate the corresponding objective function values, find the or corresponding to the minimum objective function value, update , , , , or , and repeat this process until the stop condition is met.

[0037] Among them, the process of the MBI algorithm solving the total optimization problem is shown in Table 1: Table 1 Step 104, use the SCA algorithm to solve the first sub-optimization problem and the second sub-optimization problem respectively. In each iteration of the SCA algorithm, by using a preset convex approximation function to replace the non-convex objective function and non-convex constraints in the first sub-optimization problem and the second sub-optimization problem, the first sub-optimization problem and the second sub-optimization problem are respectively transformed into the first convex optimization problem and the second convex optimization problem.

[0038] Specifically, since the first two constraints in the first sub-optimization problem and the first constraint in the second sub-optimization problem are non-convex and cannot be directly solved, in this embodiment, the SCA algorithm is used to solve the first sub-optimization problem and the second sub-optimization problem respectively. SCA is an iterative algorithm for solving non-convex optimization problems. The basic idea of this algorithm is that in each iteration, the non-convex objective function or constraint function is replaced by an appropriate convex approximation function, and then a series of convex sub-problems that are easy to solve are iteratively solved to gradually approximate the global optimal solution or local optimal solution of the original problem.

[0039] In the specific implementation, obtain the first-order Taylor expansions of the three concave functions in the first sub-optimization problem and the second sub-optimization problem; substitute the first-order Taylor expansions of the three concave functions into the first sub-optimization problem and the second sub-optimization problem respectively to transform the first sub-optimization problem and the second sub-optimization problem into the first convex optimization problem and the second convex optimization problem respectively. 、 and The first-order Taylor expansions of the three concave functions are respectively substituted into the first sub-optimization problem and the second sub-optimization problem to transform the first sub-optimization problem and the second sub-optimization problem into the first convex optimization problem and the second convex optimization problem respectively.

[0040] That is to say, the 、 constraints in the first sub-optimization problem and the constraint in the second sub-optimization problem are non-convex. For this reason, these three constraints can be approximated by corresponding convex upper bound functions (first-order Taylor expansions) to transform the non-convex problem into a convex problem.

[0041] Specifically, let and represent the solutions of the i th iteration respectively, then the first-order Taylor expansions of the three concave functions 、 and are respectively: ; ; .

[0042] Substitute these formulas into the first sub-optimization problem and the second sub-optimization problem, and the convexified sub-problems (i.e., the first convex optimization problem and the second convex optimization problem) can be obtained as follows: ; .

[0043] Step 105, solve the first convex optimization problem and the second convex optimization problem respectively to obtain the target amplitude of the transmitted waveform and the target receiving vector of the mismatched filter in the MIMO radar, so that the transmitting antenna and the mismatched filter of the MIMO radar jointly perform target detection with the target amplitude and the target receiving vector respectively.

[0044] Specifically, these two convex problems (i.e., the first convex optimization problem and the second convex optimization problem) can be directly solved by the interior point method. Once and are obtained, substitute , , and repeat this process until the stopping condition is met.

[0045] Among them, the processes of solving the first sub-optimization problem and the second sub-optimization problem by the SCA algorithm are shown in Table 2: Table 2 In this embodiment, by establishing a total optimization problem with the minimum ISL value of the MIMO radar signal GCAF as the objective under the constraints of the amplitude of the MIMO radar transmission waveform, the receiving vector of the mismatched filter, and the main lobe loss, and solving it, the transmitting antenna and the mismatched filter of the MIMO radar perform target detection with the solved target amplitude and target receiving vector respectively. This total optimization problem not only simultaneously considers the design of the radar transmission sequence set and the receiving filter, improves the radar's degrees of freedom, but also can optimize the waveform in the MIMO radar to achieve a lower ISL value and a controllable waveform peak-to-average ratio, effectively improving the accuracy of MIMO radar target detection.

[0046] In one embodiment, since the interior point algorithm is used to solve the first convex optimization problem and the second convex optimization problem in step 104 of the above embodiment, its computational complexity is about , and the amount of computation is large when N is very large. At this time, this embodiment will propose a fast algorithm based on ADPM to solve the first convex optimization problem and the second convex optimization problem.

[0047] In specific implementation, compared with the traditional ADMM algorithm, the penalty factor of the former can be dynamically updated in each iteration, making the penalty term gradually approach zero. While ensuring the convergence of the algorithm, it enables the algorithm to find a better feasible solution faster.

[0048] Therefore, based on the ADPM algorithm framework, this embodiment equivalently transforms the first convex optimization problem into: .

[0049] The augmented Lagrangian function of the equivalently transformed first convex optimization problem is: ; In the formula, and respectively represent the Lagrange multiplier vector and the penalty factor.

[0050] Adopt Indicates the optimization result at the th iteration. The optimization steps are as follows: Step 1: Update : Ignore the terms in that are irrelevant to . The optimization problem with respect to can be simplified to: ; In the formula, , , .

[0051] The KKT conditions for this formula are: ; Then, the solution to the above optimization problem with respect to can be obtained: ; In the formula, .

[0052] Step 2: Update : Ignore the terms in that are irrelevant to . The optimization problem with respect to is: ; In the formula, , .

[0053] It can be found that the objective function and constraints of this problem are separable with respect to , that is, the variables in this problem can be optimized in parallel. For the th element in , we can get:

[0054] ; Similar to the optimization problem with respect to , the closed-form solution to this formula is: ; In the formula, .

[0055] Step 3: Update : Ignore the terms in that are irrelevant to . The optimization problem with respect to The optimization problem is: ; In the formula, .

[0056] Similarly, the objective function and constraints of this problem with respect to are separable. For the th element in , we can obtain: ; Its closed-form solution is .

[0057] Step 4. Update : Ignore the terms irrelevant to and . The optimization problem with respect to is: ; In the formula, .

[0058] Further simplifying it gives: ; In the formula, .

[0059] This is an unconstrained problem. Directly taking the first derivative of the variable and setting it equal to 0, we get the solution: .

[0060] Updating according to this solution requires an inverse operation. To reduce the computational complexity, we can utilize the property of eigenvalue decomposition to avoid directly taking the inverse of in each iteration, thereby reducing the computational complexity.

[0061] The specific principle is as follows: If the singular value decomposition of is: ; In the formula, is the unitary matrix of eigenvectors, and is the diagonal matrix of eigenvalues.

[0062] Then: .

[0063] Because and remain unchanged during the iteration process of the ADPM algorithm and can be obtained in advance before the ADPM iteration starts. For subsequent updates, we only need to perform operations on Inverse. Therefore, if the maximum number of iterations is , the computational complexity can be reduced from to .

[0064] Step 5. Update and : The ADPM algorithm updates based on the original residual value . If does not decrease as the number of iterations increases, then increase to force to approach 0, so as to find a feasible solution. Otherwise, remains unchanged. Therefore is expressed as:

[0065] ; wherein, , , .

[0066] The Lagrange multiplier vector can be updated according to the following formula: ; wherein, , is a sufficiently large positive number.

[0067] Iteratively update , , , until the stopping condition is reached.

[0068] A reasonable termination condition is that both the original residual and the dual residual are small enough, that is: ; ; wherein, and are the feasibility tolerances of the original residual and the dual residual at the -th iteration respectively. These two values can be selected by absolute and relative criteria, that is: ; ; wherein, is the absolute error, is the relative error.

[0069] When both of the above reasonable termination conditions are satisfied, the ADPM iteration terminates.

[0070] Suitable It can make the ADPM algorithm converge faster. The appropriate initial penalty factor is: ; In the formula, and respectively represent the minimum and maximum eigenvalues of. If the minimum eigenvalue of is 0, then select the minimum non-zero eigenvalue of.

[0071] Among them, the process of the ADPM algorithm for solving the first convex optimization problem can be seen in Table 3: Table 3 Similarly, introduce auxiliary variables , , and equivalently transform the second convex optimization problem into: ; The augmented Lagrangian function of the equivalently transformed second convex optimization problem is: .

[0072] Since this formula has a similar structure to the augmented Lagrangian function of the equivalently transformed first convex optimization problem, the same ADPM algorithm can be used to solve it, and the specific steps are not elaborated here.

[0073] The computational complexity of this embodiment is mainly related to the number of iterations of the inner and outer layer algorithms and the sequence length . The computational complexity of the inner loop SCA algorithm is , and the computational complexity of the ADPM algorithm is . Assuming that the maximum number of iterations when the outer layer MBI algorithm converges is , then the computational complexities of the overall MBI-SCA and MBI-ADPM algorithms are respectively , . It can be seen that the complexity of the latter is generally lower than that of the former.

[0074] In this embodiment, (1) a problem model for minimizing the GCAF autocorrelation and cross-correlation ISL under the main lobe gain and dynamic range constraints is established. This model can not only provide a lower ISL value, but also control the peak-to-average ratio of the waveform, so that the designed waveform can better apply to the actual MIMO system.

[0075] (2) Aiming at the HOP non-convex problem, the MBI_SCA method is proposed. By using the MBI algorithm, the non-convex problem is decomposed into multiple parallel sub-problems, and then the SCA algorithm is used to iteratively solve the sub-problems.

[0076] (3) To reduce the computational complexity, a fast ADPM algorithm with the potential for parallel implementation is proposed to solve the sub-problem of SCA, which ensures convergence while improving the computational efficiency and can obtain a lower ISL in some cases.

[0077] In one embodiment, the effectiveness of the method proposed in the above embodiment can be verified by MATLAB numerical simulation in the present invention.

[0078] Among them, the simulation parameters are set as follows: The initial transmission sequence and the receiving filter used in the experiment are both all-1 sequences. The number of transmitting and receiving antennas , the sequence length , the range-Doppler cell , , , , , . The maximum number of iterations of the outer MBI algorithm , the maximum number of iterations of the inner SCA algorithm , the maximum number of iterations of the ADPM algorithm .

[0079] The simulation data results and analysis are as follows: Figure 2 is a curve graph showing the change of the objective function values of the MBI-SCA and MBI-ADPM algorithms with the number of iterations. It can be seen from Figure 2 that the MBI-SCA algorithm converges after about 12 iterations, while the MBI-ADPM algorithm converges after about 10 iterations, with a faster convergence speed and can reach a lower objective function value .

[0080] Figure 3 is a three-dimensional graph of the ISL of the range-Doppler cell obtained by the MBI-SCA method, Figure 4 is a three-dimensional graph of the ISL of the range-Doppler cell obtained by the MBI-ADPM method, and the top is all , , and the bottom is , . It can be seen from Figure 3 , Figure 4 that both algorithms have good optimization capabilities, making the ISL values in the locally optimized region much lower than those in the unoptimized region and both less than -200 dB;

[0081] Figure 5 is a zero-Doppler two-dimensional graph. It can be seen from Figure 5It can be seen that MBI-ADPM has better local optimization ability than the MBI-SCA algorithm, which can make the local GCAFs value reach -320 dB, while MBI-SCA can only reach -260 dB.

[0082] The step division of the above various methods is only for clear description. When implemented, they can be combined into one step or some steps can be split into multiple steps. As long as the same logical relationship is included, they are all within the protection scope of the present invention; adding insignificant modifications or introducing insignificant designs to the algorithm or process, but without changing the core design of its algorithm and process, are all within the protection scope of this invention.

[0083] Another embodiment of the present invention relates to a target detection system for a MIMO radar. The implementation details of the target detection system for the MIMO radar in this embodiment will be specifically described below. The following content is only the implementation details provided for convenient understanding and is not necessary for implementing this solution. The target detection system for the MIMO radar in this embodiment includes: A parameter acquisition module, configured to acquire a transmission waveform set composed of transmission waveforms of multiple transmitting antennas, a mismatched filtering vector set composed of receiving vectors of multiple mismatched filters, and main lobe loss in the MIMO radar; A problem establishment module, configured to establish an overall optimization problem with the goal of minimizing the ISL value of the generalized cross ambiguity function (GCAF) of the MIMO radar signal under the constraints of the amplitude of the transmission waveform, the receiving vector of the mismatched filter, and the main lobe loss; A problem decomposition module, configured to solve the overall optimization problem by using the MBI algorithm. In each iteration of the MBI algorithm, by fixing the variables other than the amplitude of the transmission waveform in the overall optimization problem, or fixing the variables other than the receiving vector of the fixed mismatched filter in the overall optimization problem, the overall optimization problem is decomposed into a first sub-optimization problem with the goal of minimizing the ISL value of the GCAF of the MIMO radar signal and optimizing the receiving vector of the mismatched filter, and a second sub-optimization problem with the goal of minimizing the ISL value of the GCAF of the MIMO radar signal and optimizing the amplitude of the transmission waveform; A problem transformation module, configured to solve the first sub-optimization problem and the second sub-optimization problem respectively by using the SCA algorithm. In each iteration of the SCA algorithm, by using a preset convex approximation function to replace the non-convex objective function and non-convex constraints in the first sub-optimization problem and the second sub-optimization problem, the first sub-optimization problem and the second sub-optimization problem are respectively transformed into a first convex optimization problem and a second convex optimization problem; A problem-solving module is configured to solve the first convex optimization problem and the second convex optimization problem respectively, so as to obtain the target amplitude of the transmission waveform and the target receiving vector of the mismatch filter in the MIMO radar, so that the transmitting antenna and the mismatch filter of the MIMO radar jointly perform target detection with the target amplitude and the target receiving vector respectively.

[0084] It is not difficult to find that this embodiment is a system embodiment corresponding to the above method embodiment, and this embodiment can be implemented in cooperation with the above method embodiment. The relevant technical details and technical effects mentioned in the above embodiments are still valid in this embodiment. To avoid repetition, they will not be elaborated here. Correspondingly, the relevant technical details mentioned in this embodiment can also be applied to the above embodiments.

[0085] It is worth mentioning that each module involved in this embodiment is a logical module. In practical applications, a logical unit can be a physical unit, a part of a physical unit, or a combination of multiple physical units. In addition, to highlight the innovative part of the present invention, units that are not closely related to solving the technical problems proposed by the present invention are not introduced in this embodiment, but this does not mean that there are no other units in this embodiment.

[0086] Another embodiment of the present invention relates to a computer device, including: at least one processor; and a memory communicatively connected to the at least one processor; wherein, the memory stores instructions executable by the at least one processor, and the instructions are executed by the at least one processor, so that the at least one processor can execute the low-ISL target detection method in the above embodiments.

[0087] Wherein, the memory and the processor are connected by a bus. The bus can include any number of interconnected buses and bridges, and the bus connects various circuits of one or more processors and the memory together. The bus can also connect various other circuits such as peripheral devices, voltage regulators, and power management circuits, which are well known in the art, so they will not be further described herein. The bus interface provides an interface between the bus and the transceiver. The transceiver can be one element or multiple elements, such as multiple receivers and transmitters, and provides a unit for communicating with various other devices on the transmission medium. The data processed by the processor is transmitted over the wireless medium through the antenna. Further, the antenna also receives data and transmits the data to the processor.

[0088] The processor is responsible for managing the bus and general processing, and can also provide various functions, including timing, peripheral interface, voltage regulation, power management, and other control functions. The memory can be used to store the data used by the processor when performing operations.

[0089] Another embodiment of the present invention relates to a computer-readable storage medium storing a computer program. When the computer program is executed by a processor, the method embodiments described above are implemented.

[0090] That is, those skilled in the art can understand that all or part of the steps in implementing the methods of the above embodiments can be completed by instructing relevant hardware through a program. The program is stored in a storage medium and includes several instructions to enable a device (which can be a single-chip microcomputer, a chip, etc.) or a processor to execute all or part of the steps of the methods described in various embodiments of the present invention. The foregoing storage medium includes: various media such as USB flash drives, mobile hard disks, read-only memories (ROM), random access memories (RAM), magnetic disks, or optical discs that can store program codes.

[0091] Those of ordinary skill in the art can understand that the above embodiments are specific embodiments for implementing the present invention, and in practical applications, various changes can be made to them in form and details without departing from the spirit and scope of the present invention.

Claims

1. A target detection method for MIMO radar, characterized in that: The method comprises: Acquire a transmission waveform set composed of transmission waveforms of multiple transmission antennas, a mismatched filter vector set composed of reception vectors of multiple mismatched filters, and a main lobe loss in a MIMO radar; Under the constraints of the amplitude of the transmitted waveform, the receiving vector of the mismatched filter and the main lobe loss, a total optimization problem is established with the goal of minimizing the ISL value of the generalized mutual ambiguity function GCAF of the MIMO radar signal. The MBI algorithm is used to solve the total optimization problem. In each iteration of the MBI algorithm, by fixing variables other than the amplitude of the transmitted waveform in the total optimization problem, or fixing variables other than the receiving vector of the mismatched filter in the total optimization problem, the total optimization problem is decomposed into a first sub-optimization problem for optimizing the receiving vector of the mismatched filter with the goal of minimizing the ISL value of the MIMO radar signal GCAF, and a second sub-optimization problem for optimizing the amplitude of the transmitted waveform with the goal of minimizing the ISL value of the MIMO radar signal GCAF. The SCA algorithm is used to solve the first sub-optimization problem and the second sub-optimization problem respectively. In each iteration of the SCA algorithm, the first sub-optimization problem and the second sub-optimization problem are respectively transformed into the first convex optimization problem and the second convex optimization problem by using a preset convex approximation function to replace the non-convex objective function and non-convex constraints in the first sub-optimization problem and the second sub-optimization problem; The first convex optimization problem and the second convex optimization problem are solved respectively to obtain the target amplitude of the transmitting waveform in the MIMO radar and the target receiving vector of the mismatched filter, so that the transmitting antenna and the mismatched filter of the MIMO radar can jointly detect the target with the target amplitude and the target receiving vector respectively.

2. The target detection method of MIMO radar according to claim 1, characterized in that: The GCAF of the MIMO radar signal is: ； In the formula, Indicates The set of transmit waveforms for the transmit antennas, Indicates The set of mismatched filter vectors for the receiving antennas, represents the length of the transmitted waveform set, represents the Doppler frequency shift vector set, represents the delay set, N -1= K ; Simplified expression: ； In the formula, represents the conjugate transpose operation, Represented by vector The diagonal matrix formed by and The matrices representing the Doppler frequency shift vector and time delay respectively; ； ； The overall optimization problem is expressed by the following formula: ； Where C1, C2 and C3 represent the main lobe loss constraint, the amplitude constraint of the transmit waveform and the receive vector constraint of the mismatch filter, respectively. represents the preset main lobe gain coefficient, represents the main lobe set, Indicates the fluctuation degree of the transmitted waveform amplitude. represents the set of all side lobes of the local GCAF.

3. The target detection method of MIMO radar according to claim 2, characterized in that: The first sub-optimization problem and the second sub-optimization problem are respectively expressed by the following formulas: ； 。 4. The target detection method of MIMO radar according to claim 3, characterized in that: The solving of the first sub-optimization problem and the second sub-optimization problem respectively comprises: Get the first sub-optimization problem and the second sub-optimization problem , and First-order Taylor expansions of three concave functions; The first-order Taylor expansions of the three concave functions are substituted into the first sub-optimization problem and the second sub-optimization problem, respectively, so as to transform the first sub-optimization problem and the second sub-optimization problem into the first convex optimization problem and the second convex optimization problem, respectively.

5. The target detection method of MIMO radar according to claim 4, characterized in that: The first convex optimization problem and the second convex optimization problem are respectively expressed by the following formulas: ； 。 6. The target detection method of MIMO radar according to claim 5, characterized in that: The first convex optimization problem is solved by the following steps: For the first convex optimization problem, an auxiliary variable is introduced , and , the first convex optimization problem is equivalent to: ； The augmented Lagrangian function of the equivalent first convex optimization problem is: ； In the formula, and denote the Lagrange multiplier vector and penalty factor respectively; based on For each item in Optimization problem about Optimization problem about The optimization problem and the The optimization problem is: ； In the formula, , , ; ； In the formula, , ; ； In the formula, ; ； In the formula, ; Based on the ADPM algorithm, Optimization problem about Optimization problem about The optimization problem and the The optimization problem is solved so that the penalty factor in the ADPM algorithm is dynamically updated in each iteration, and the penalty term gradually approaches zero.

7. A target detection system for a MIMO radar, characterized in that: The system comprises: A parameter acquisition module is used to acquire a transmission waveform set composed of transmission waveforms of multiple transmission antennas, a mismatch filter vector set composed of receiving vectors of multiple mismatch filters, and a main lobe loss in a MIMO radar; A problem establishment module is used to establish a total optimization problem with the goal of minimizing the ISL value of the generalized mutual ambiguity function GCAF of the MIMO radar signal under the constraints of the amplitude of the transmitted waveform, the receiving vector of the mismatched filter and the main lobe loss; A problem decomposition module is used to solve the overall optimization problem by using an MBI algorithm. In each iteration of the MBI algorithm, by fixing variables other than the amplitude of the transmitted waveform in the overall optimization problem, or fixing variables other than the received vector of the mismatched filter in the overall optimization problem, the overall optimization problem is decomposed into a first sub-optimization problem for optimizing the received vector of the mismatched filter with the goal of minimizing the ISL value of the MIMO radar signal GCAF, and a second sub-optimization problem for optimizing the amplitude of the transmitted waveform with the goal of minimizing the ISL value of the MIMO radar signal GCAF. A problem conversion module, used for respectively solving the first sub-optimization problem and the second sub-optimization problem by using the SCA algorithm, and in each iteration of the SCA algorithm, by replacing the non-convex objective function and non-convex constraints in the first sub-optimization problem and the second sub-optimization problem with a preset convex approximation function, the first sub-optimization problem and the second sub-optimization problem are respectively converted into a first convex optimization problem and a second convex optimization problem; The problem solving module is used to solve the first convex optimization problem and the second convex optimization problem respectively, and obtain the target amplitude of the transmitting waveform in the MIMO radar and the target receiving vector of the mismatch filter, so that the transmitting antenna and the mismatch filter of the MIMO radar can jointly detect the target with the target amplitude and the target receiving vector respectively.

8. A computer device, characterized in that: include: at least one processor; And, a memory communicatively connected to the at least one processor; wherein the memory stores instructions executable by the at least one processor, and the instructions are executed by the at least one processor so that the at least one processor can execute the target detection method of the MIMO radar as described in any one of claims 1 to 6.

9. A computer-readable storage medium storing a computer program, characterized in that: When the computer program is executed by a processor, the target detection method of the MIMO radar according to any one of claims 1 to 6 is implemented.

Images