MIMO sparse array optimization method and device based on genetic algorithm, medium and product

Abstract

The invention provides an MIMO sparse array optimization method and device based on a genetic algorithm, a medium and a product. The method comprises the steps that ROE is selected based on the physical size of a radar board; coding the physical array of the transmitting and receiving antenna according to ROE to generate an initial population; generating a virtual array population from the initial population through an MIMO equivalent phase center; angle measurement array elements in the horizontal direction and the pitching direction are extracted for the virtual array population; a normalized fitness function index is constructed for a sub-array formed after array forming of angle measurement array elements in the horizontal direction and the pitching direction, and the individual fitness after array forming is calculated; for the current population, selecting antenna array arrangement individuals for subsequent operation; carrying out individual position crossing; performing individual position variation; and when a termination condition is satisfied, outputting an optimal individual. By using the above technical scheme, the MIMO physical array arrangement is optimized, the premature problem of the genetic algorithm can be suppressed, the angular resolution of the radar is improved, and the result after sparse array optimization is improved.

Description

Technical Field

[0001] The present invention relates to the field of radar array signal processing, and in particular to a method, device, medium and product for optimizing a MIMO sparse array based on a genetic algorithm. Background Art

[0002] Angular resolution is a key performance indicator in MIMO radar systems. Existing technologies improve angular resolution by increasing the number of array elements. However, this will lead to a significant increase in the integration difficulty and cost of MIMO radar systems. The array sparse design method can achieve the required beam width and angular resolution with fewer array elements, thereby saving hardware costs and reducing processing complexity; this method is an important way to effectively solve the problem of mutual restriction between angular resolution performance and MIMO array cost. In most works, a lot of efforts have been made in the design of MIMO arrays, but the distribution of MIMO radar virtual elements is mainly focused on, resulting in many array layout results after sparse optimization that cannot be realized by physical element layout. Moreover, the existing intelligent optimization algorithms based on the array sparse design method have the problem of premature convergence in the process of finding the optimal solution.

[0003] Specifically, the working logic of the existing array optimization method based on genetic algorithm includes:

[0004] Generating an initial array population by encoding according to the number of virtual elements;

[0005] Calculating the peak sidelobe level (PSLL) according to the pattern model as the fitness function value;

[0006] Performing individual selection according to the genetic algorithm;

[0007] Performing individual position crossover according to the crossover rate formula;

[0008] Performing individual position mutation according to the mutation rate formula;

[0009] Judging whether the fitness of the propagated individual meets the requirements or the number of iterations reaches the upper limit;

[0010] Outputting the optimal virtual element position.

[0011] The existing above-mentioned array optimization method has the following disadvantages:

[0012] The generation of the initial array population is all single linear arrays or planar arrays, without considering the physical array layout scale of the actual radar board;

[0013] Using PSLL as the fitness evaluation index according to the pattern model is too single, resulting in the problem of premature convergence in the process of the optimization algorithm finding the optimal solution;

[0014] The optimized output of the virtual array element positions fails to directly match the actual physical array, which will render the optimization result ineffective. Summary of the Invention

[0015] In view of one or more of the above problems of the prior art, embodiments of the present invention provide a MIMO sparse array optimization method, device, medium and product based on a genetic algorithm.

[0016] To achieve the above object, on the one hand, a MIMO sparse array optimization method based on a genetic algorithm is provided, including:

[0017] Select an executable area based on the predetermined radar field of view requirements, the number of antennas, and the physical size of the radar board, where the executable area is used to initialize the arrangement of the transmitting and receiving antennas;

[0018] Encode the physical array of the transmitting and receiving antennas according to the executable area to generate an initial population;

[0019] Generate a virtual array population from the initial population through the MIMO equivalent phase center;

[0020] Extract the angle measurement array elements in the horizontal and elevation directions for the virtual array population;

[0021] Construct a normalized fitness function index for the subarray formed after shaping the angle measurement array elements in the horizontal and elevation directions, and calculate the individual fitness after array shaping according to the fitness function index, where the fitness function index is defined by the following formula:

[0022] Fitness function index = W 1 *(W 11 *az_PSLL + W 12 *az_Antennanum + W 13 *az_Aperpure) + W 2 *(W 21 *el_PSLL + W 22 *el_Antennanum + W 23 *az_Aperpure);

[0023] Wherein, W 1 + W 2 = 1, W 11 + W 12 + W 13 = 1, W 21 + W 22 + W 23 = 1;

[0024] Wherein, W 1 is the weight in the horizontal direction, W2 is the weight in the pitch direction, W 11 , W 12 and W 13 are the weights corresponding to the peak sidelobe level az_PSLL in the horizontal direction, the number of array elements az_Antennanum, and the array aperture az_Aperpure respectively, W 21 , W 22 and W 23 are the weights corresponding to the peak sidelobe level el_PSLL in the pitch direction, the number of array elements el_Antennanum, and the array aperture el_Aperpure respectively;

[0025] For the current population, select the individual of the antenna array arrangement for subsequent operations in the form of roulette wheel;

[0026] Perform individual position crossover according to the predetermined crossover rate formula;

[0027] Perform individual position mutation according to the preset mutation rate formula;

[0028] When the current optimal fitness function index is greater than the predetermined target fitness function index value, output the optimal individual.

[0029] Preferably, for the MIMO sparse array optimization method, the physical array of the transceiver antennas is encoded as: the positions of the physical array of the transceiver antennas are binary encoded using 0 and 1; wherein, 0 indicates that there is no antenna at the corresponding position; 1 indicates that there is an antenna at the corresponding position.

[0030] Preferably, for the MIMO sparse array optimization method, generating the virtual array population from the initial population through the MIMO equivalent phase center includes:

[0031] For each individual in the initial population, N t transmitting array elements and N r receiving array elements are equivalent to N t *N r transceiver - shared virtual array elements according to the equivalent phase center principle.

[0032] Preferably, for the MIMO sparse array optimization method, extracting the angle - measuring array elements in the horizontal and pitch directions for the virtual array population includes:

[0033] Taking the maximum aperture length as an index, shaping the horizontal - direction guiding vector array by retrieving the horizontal - direction maximum aperture length array; and,

[0034] Taking the maximum aperture length as an index, shaping the pitch - direction guiding vector array by retrieving the pitch - direction maximum aperture length array.

[0035] Preferably, for the MIMO sparse array optimization method, the weights W 1 , W 2 , W 11 , W 12 , W 13 , W 21 , W 22 and W 23 are preset.

[0036] On the other hand, a MIMO sparse array optimization device based on a genetic algorithm is provided, including a memory and a processor. The memory stores at least one program, and the at least one program is executed by the processor to implement the steps of the MIMO sparse array optimization method based on a genetic algorithm as described above in any one of the preceding paragraphs.

[0037] On another aspect, a computer-readable storage medium is provided. The storage medium stores at least one program, and the at least one program is executed by the processor to implement the steps of the MIMO sparse array optimization method based on a genetic algorithm as described above in any one of the preceding paragraphs.

[0038] On another aspect, a computer program product is provided, including a computer program. It is characterized in that when the computer program is executed by a processor, it implements the steps of the MIMO sparse array optimization method based on a genetic algorithm as described above in any one of the preceding paragraphs.

[0039] On another aspect, a radar is provided, and the MIMO sparse array optimization method based on a genetic algorithm as described above in any one of the preceding paragraphs is used for MIMO sparse array optimization.

[0040] On another aspect, a vehicle is provided, including the radar as described above.

[0041] The above technical solutions have the following technical effects:

[0042] The technical solution of the embodiment of the present invention selects an executable area based on the actual physical size of the radar to initialize the arrangement of the transceiver antennas, that is, to initialize the element positions, ensuring that the maximum length of the array obtains the optimal element positions, thereby effectively reducing the cost and calculation amount of the antenna system, and better meeting the requirements of actual engineering design; moreover, the technical solution of the embodiment of the present invention optimizes the MIMO physical array layout by extracting the azimuth and elevation angle measurement elements for the virtual array population and using the combined azimuth and elevation to construct the fitness evaluation index of the comprehensive genetic algorithm, suppressing the premature problem of the genetic algorithm, improving the angular resolution of the radar and the result after sparse array optimization. Description of the Drawings

[0043] Figure 1Schematic flow chart of the MIMO sparse array optimization method based on genetic algorithm according to an embodiment of the present invention;

[0044] Figure 2 Schematic diagram of an exemplary ROE used in the MIMO sparse array optimization method based on genetic algorithm according to an embodiment of the present invention;

[0045] Figure 3 Schematic diagram of an exemplary initial population generated in the MIMO sparse array optimization method based on genetic algorithm according to an embodiment of the present invention;

[0046] Figure 4 is Figure 2 Schematic diagram of the virtual receiving array formed by the physical antenna array in the ROE shown;

[0047] Figure 5 Schematic diagram of an exemplary angle measuring element in the horizontal and elevation directions;

[0048] Figure 6 Schematic diagram of the positions of an exemplary MIMO radar physical antenna and receiving antenna;

[0049] Figure 7 Schematic diagram of an exemplary target coordinate system;

[0050] Figure 8 Schematic diagram of the basic process of the genetic algorithm. Detailed implementation manners

[0051] To further illustrate the embodiments, the present invention provides drawings. These drawings are part of the disclosure of the present invention, mainly used to illustrate the embodiments, and can be combined with the relevant descriptions in the specification to explain the operating principles of the embodiments. With reference to these contents, those of ordinary skill in the art should be able to understand other possible implementation manners and the advantages of the present invention. The components in the drawings are not drawn to scale, and similar component symbols are usually used to represent similar components.

[0052] The present invention will be further described below in conjunction with the drawings and specific implementation manners.

[0053] Embodiment 1:

[0054] Figure 1 Schematic flow chart of the MIMO sparse array optimization method based on genetic algorithm according to an embodiment of the present invention. As Figure 1 , the MIMO sparse array optimization method based on genetic algorithm in this embodiment includes:

[0055] S1. Select the Region of Executable (ROE) based on the predetermined radar field of view requirements, the number of antennas, and the physical size of the radar board. The ROE is used to initialize the arrangement of the transceiver antennas.

[0056] In a specific implementation, denote the carrier wavelength of the radar transmitted signal as λ, the minimum element spacing is in units of half wavelength, and the number of transmitting elements is N. t The number of receiving elements is N. r Select the ROE according to the field of view (FOV) requirements during the actual application of the radar, the number of antennas, and the physical size of the radar board.

[0057] Figure 2 Figure 11 is an exemplary schematic diagram of the ROE. Where Tx is the transmitting antenna and Rx is the receiving antenna; dX is the length of the ROE, and dZ is the width of the ROE. Figure 2 Figure 13 shows a 4 - transmit and 4 - receive physical antenna array in the ROE.

[0058] S2. Encode the physical array of the transceiver antennas according to the ROE to generate an initial population.

[0059] In a specific implementation, the encoding of the physical array of the transceiver antennas is as follows: Use 0 and 1 for binary encoding of the positions of the physical array of the transceiver antennas, that is, use 0 and 1 at the positions of the transceiver antennas to indicate the presence or absence of the antenna at the corresponding position. 0 indicates that there is no antenna at the corresponding position; 1 indicates that there is an antenna at the corresponding position; all 0s and 1s constitute the initial population. Figure 3 Figure 21 is an exemplary schematic diagram of the initial population. Figure 3 It includes the individual length Indivlength of the population and the initial population size Popsize. The size of the initial population P(t) affects the result of subsequent iterative optimization. Briefly speaking, the larger the initial population, the easier it is to obtain the target individual in subsequent iterative optimization, but the corresponding computational complexity also increases.

[0060] S3. Generate a virtual array population from the initial population through the MIMO equivalent phase center.

[0061] In a specific implementation, for each individual in the initial population, N t transmitting elements and N r receiving elements are equivalent to a virtual element shared by both transmitting and receiving according to the principle of the equivalent phase center. Denote K = N t *N r to represent the number of virtual elements. Figure 4 Figure 41 is Figure 2 a schematic diagram of the virtual receiving array VRX formed by the 4 - transmit and 4 - receive physical antenna array shown in Figure 40.

[0062] S4. Extract the angle measurement array elements in the horizontal and pitch directions for the virtual array population;

[0063] Since most actual vehicle-mounted radars are limited by hardware processors during operation, the processing link for actual direction-of-arrival (DOA) angle measurement is to perform pitch DOA after separate horizontal DOA; therefore, in a specific implementation of the embodiments of the present invention, extracting the angle measurement array elements in the horizontal and pitch directions for the virtual array population includes:

[0064] Taking the maximum aperture length as an index, shaping the horizontal guiding vector array by retrieving the horizontal maximum aperture length array; and, taking the maximum aperture length as an index, shaping the pitch guiding vector array by retrieving the pitch maximum aperture length array. Figure 5 FIG. is a schematic diagram of an exemplary angle measurement array element in the horizontal and pitch directions.

[0065] S5. Construct a normalized fitness function index for the sub-array formed after shaping the angle measurement array elements in the horizontal and pitch directions, and calculate the individual fitness after array shaping according to the fitness function index, where the fitness function index is defined by the following formula:

[0066] Fitness function index = W 1 *(W 11 *az_PSLL + W 12 *az_Antennanum + W 13 *az_Aperpure) + W 2 *(W 21 *el_PSLL + W 22 *el_Antennanum + W 23 *az_Aperpure);

[0067] where, W 1 + W 2 = 1, W 11 + W 12 + W 13 = 1, W 21 + W 22 + W 23 = 1;

[0068] where, W 1 is the weight in the horizontal direction, W 2 is the weight in the pitch direction, W 11 , W 12 and W 13 are the weights corresponding to the peak sidelobe level az_PSLL, the number of array elements az_Antennanum, and the array aperture az_Aperpure in the horizontal direction respectively, W 21, W 22 and W 23 are the weights corresponding to the peak sidelobe level el_PSLL, the number of array elements el_Antennanum, and the array aperture el_Aperpure in the pitch direction, respectively;

[0069] In a specific implementation, according to the pre-set weight W 1 , W 2 , W 11 , W 12 , W 13 , W 21 , W 22 and W 23 values; and the values of the weights can be adjusted according to actual needs to make the algorithm iteration optimization more flexible to solve the problem of individual premature convergence.

[0070] S6. For the current population, use the roulette wheel method to select the antenna array layout individuals for subsequent operations;

[0071] In a specific implementation, the roulette wheel method is used to perform multiple rounds of selection operations for the current population; the roulette wheel method is used to select the antenna array layout individuals for subsequent operations for the initial population; the roulette wheel is a proportion-based selection that uses the proportion of individual fitness to determine the possibility of offspring retention. The greater the individual fitness, the greater the chance of being selected.

[0072] S7. Perform individual position crossover according to the pre-set crossover rate formula;

[0073] Crossover is to randomly pair the selected individuals in the population P(t). For each pair of individuals, exchange some of their genes with a pre-set crossover probability, such as the crossover probability Pc = 0.8, to form new individuals through crossover, and then form a new population with the new individuals;

[0074] S8. Perform individual position mutation according to the pre-set mutation rate formula;

[0075] Perform a probability-based mutation operation on each array individual in the new population generated after crossover. Under the condition of ensuring the individual sparsity rate, according to the pre-set mutation probability, such as Pm = 0.05, change the gene values at some loci of each individual to other allele values. In the case of binary coding, just invert the corresponding gene values;

[0076] S9. When the current optimal fitness function index is greater than the pre-set target fitness function index value, output the optimal individual;

[0077] In this step, if the current optimal fitness function index is not greater than the predetermined target fitness function index value, then jump back to step S3 for iterative optimization until the target fitness function index value is satisfied;

[0078] The technical solution of the embodiment of the present invention uses the actual radar physical size to construct the executable area, uses the actual physical antenna position as the initial population for optimization, introduces the equivalent antenna phase principle to analyze the MIMO virtual array, constructs a comprehensive fitness index by combining azimuth (i.e., horizontal direction) and elevation, and reconstructs the genetic algorithm process to optimize the iterative result, optimizes the MIMO physical array layout, suppresses the premature problem of the genetic algorithm, improves the angular resolution of the radar, improves the layout rationality of the physical array, and improves the result after sparse array optimization.

[0079] Next, the basic principles of MIMO equivalent phase center, array shaping, and genetic algorithm involved in the MIMO sparse array optimization method of the embodiment of the present invention will be described.

[0080] 1. MIMO Equivalent Phase Center

[0081] Figure 6 It is a schematic diagram of the positions of the physical antennas and receiving antennas of an exemplary MIMO radar. Figure 6 Taking 4 transmit and 4 receive as an example.

[0082] As Figure 6 , the position sets of the physical transmit antennas and receive antennas of the MIMO radar are:

[0083]

[0084] Among them, N t and N r are the numbers of physical transmit antennas and receive antennas respectively; P Tx is the position set of the horizontal transmit antennas, and P Tz is the position set of the elevation transmit antennas; P Rx is the position set of the horizontal receive antennas, and P Rz is the position set of the elevation receive antennas.

[0085] As Figure 7 shown in the target coordinate system, the steering vector of the target DOA associated transmit signal and the steering vector of the receive signal can be respectively expressed as:

[0086]

[0087] Among them, θ is the elevation angle, is the horizontal angle.

[0088] Therefore, the virtual steering vector can be expressed as the Kronecker product of the steering vector of the transmitted signal and the steering vector of the received signal:

[0089]

[0090] The virtual steering vector in array form is:

[0091]

[0092] Therefore, the set of virtual array elements formed by the MIMO equivalent phase center principle is:

[0093]

[0094]

[0095] where P vx is the set of virtual array elements in the horizontal direction, and P vz is the set of virtual array elements in the elevation direction.

[0096] The maximum array apertures d x and d z of the virtual array in the horizontal and elevation directions can be defined as:

[0097] d x = max(P vx ) - min(P vx )

[0098] d z = max(P vz ) - min(P vz )

[0099] 2. Array shaping

[0100] Considering that the maximum value of the array beam points to and the antenna illumination aperture is an equal-amplitude omnidirectional distribution, according to the above steering vector representation, the antenna array pattern function can be expressed as:

[0101]

[0102] where λ is the radar wavelength, N and M are the number of arrays in the horizontal and elevation directions, d m is the element spacing in the horizontal direction, and d n is the element spacing in the elevation direction. Therefore, the maximum sidelobe level of the antenna pattern can be expressed as:

[0103]

[0104] Among them, is expressed as the antenna array pattern function, S is the sidelobe range of the pattern, and PSLL is the key index for optimizing the sparse array; is the antenna array pattern function expressed in dB;

[0105] 3. Genetic Algorithm

[0106] The main idea of the genetic algorithm is borrowed from Darwin's evolutionary model under natural selection. By borrowing the biological evolution theory, the genetic algorithm simulates the problem to be solved as a process of biological gene genetic evolution. Through operations such as replication, crossover, and mutation, the solutions of the next generation are generated, and the solutions with low fitness function values are gradually eliminated, while the solutions with high fitness function values are increased. Figure 8 is the schematic diagram of the basic process of the genetic algorithm.

[0107] Such as Figure 8 , the basic process of the genetic algorithm in the prior art includes:

[0108] Encoding and generating the initial population P(t);

[0109] Calculating the individual fitness value of each individual in the initial population P(t);

[0110] Selection;

[0111] Crossover;

[0112] Mutation;

[0113] Obtaining the next population P(t + 1), decoding, calculating the fitness, and continuing to transfer to selection until the iteration termination condition is met, and then ending.

[0114] Embodiment 2:

[0115] The present invention also provides a MIMO sparse array optimization device based on a genetic algorithm. The device includes a processor and a memory. The memory stores at least one program, and the at least one program is executed by the processor to implement the steps of the MIMO sparse array optimization method embodiment as described above based on the genetic algorithm.

[0116] Embodiment 3:

[0117] The present invention also provides a computer-readable storage medium. The computer-readable storage medium stores a computer program, and when the computer program is executed by a processor, it implements the steps of the method in the embodiments of the present invention as described above.

[0118] If the modules / units integrated in the computer unit are implemented in the form of software functional units and sold or used as independent products, they can be stored in a computer-readable storage medium. Based on this understanding, to implement all or part of the processes in the above-described embodiment methods of the present invention, it can also be completed by instructing relevant hardware through a computer program. The computer program can be stored in a computer-readable storage medium. When the computer program is executed by a processor, the steps of the above-described method embodiments can be implemented. Among them, the computer program includes computer program code, and the computer program code can be in the form of source code, object code, executable file, or some intermediate form, etc. The computer-readable medium can include: any entity or device capable of carrying the computer program code, recording medium, USB flash drive, mobile hard disk, magnetic disk, optical disk, computer memory, read-only memory (ROM, Read-Only Memory), random access memory (RAM, Random Access Memory), and software distribution medium, etc. It should be noted that the content included in the computer-readable medium can be appropriately increased or decreased according to the requirements of legislation and patent practice in the jurisdiction.

[0119] Embodiment Four:

[0120] The present invention also provides a computer program product, including a computer program, and when the computer program is executed by a processor, the steps of the method as described above are implemented.

[0121] Embodiment Five:

[0122] The present invention also provides a radar, which uses the MIMO sparse array optimization method based on genetic algorithm as described in any of the above to optimize the MIMO sparse array.

[0123] Embodiment Six:

[0124] The present invention also provides a vehicle, including the radar as described above.

[0125] Although the present invention has been specifically shown and described in conjunction with the preferred embodiments, those skilled in the art should understand that various changes can be made to the present invention in terms of form and details without departing from the spirit and scope of the present invention defined by the appended claims, and all of them fall within the protection scope of the present invention.

Claims

1. A MIMO sparse array optimization method based on genetic algorithm, characterized in that: include: Selecting an executable area based on a predetermined radar field of view angle requirement, the number of antennas, and the physical size of the radar board, wherein the executable area is used to initialize the arrangement of the transceiver antennas; Encoding the physical array of the transceiver antennas according to the executable area to generate an initial population; The initial population is used to generate a virtual array population through a MIMO equivalent phase center; Extracting horizontal and elevation angle measurement array elements for the virtual array population; A normalized fitness function index is constructed for the sub-array formed by the angular measurement array element arrays in the horizontal direction and the pitch direction after shaping, and the individual fitness of the array after shaping is calculated according to the fitness function index, wherein the fitness function index is defined by the following formula: Fitness function index = W1*(W 11 *az_PSLL+W 12 *az_Antennanum+W 13 *az_Aperpure)+W2*(W 21 *el_PSLL+W 22 *el_Antennanum+W 23 *el_Aperpure); Among them, W1 + W2 = 1, W 11 + W 12 + W 13 = 1, W 21 + W 22 + W 23 = 1; Among them, W1 is the horizontal weight, W2 is the pitch weight, and W 11 , W 12 and W 13 are the weights corresponding to the horizontal peak sidelobe level az_PSLL, the number of array elements az_Antennanum and the array aperture az_Aperpure, respectively. 21 , W 22 and W 23 They are the weights corresponding to the peak sidelobe level el_PSLL in elevation, the number of array elements el_Antennanum and the array aperture el_Aperpure; For the current population, use roulette to select the antenna array arrangement individuals for subsequent operations; Individual positions are crossed according to a predetermined crossover rate formula; Perform individual position mutation according to a pre-set mutation rate formula; When the current optimal fitness function index is greater than the predetermined target fitness function index value, the optimal individual is output.

2. The MIMO sparse array optimization method according to claim 1, characterized in that: The physical array of the transceiver antenna is encoded as follows: the position of the physical array of the transceiver antenna is binary-encoded using 0 and 1; wherein 0 indicates that there is no antenna at the corresponding position; and 1 indicates that there is an antenna at the corresponding position.

3. The MIMO sparse array optimization method according to claim 1, characterized in that: Generating a virtual array population from the initial population through the MIMO equivalent phase center includes: For each individual in the initial population, N t transmit array elements and N r The receiving array elements are equivalent to N t *N r A virtual array element shared by both transmitter and receiver.

4. The MIMO sparse array optimization method according to claim 1, characterized in that: Extracting the angle measurement array elements in the horizontal direction and the pitch direction for the virtual array population includes: Taking the maximum aperture length as an index, shaping the horizontal steering vector array by retrieving the horizontal maximum aperture length array; and, Taking the maximum aperture length as an indicator, the pitch steering vector array is shaped by retrieving the maximum aperture length array in the pitch direction.

5. The MIMO sparse array optimization method according to claim 1, characterized in that: The weights W1, W2, W 11 , W 12 , W 13 , W 21 , W 22 and W 23 The value is preset.

6. A MIMO sparse array optimization device based on genetic algorithm, characterized in that: It comprises a memory and a processor, wherein the memory stores at least one program, and the at least one program is executed by the processor to implement the steps of the MIMO sparse array optimization method based on a genetic algorithm as claimed in any one of claims 1 to 5.

7. A computer-readable storage medium, characterized in that: The storage medium stores at least one program, and the at least one program is executed by a processor to implement the steps of the MIMO sparse array optimization method based on a genetic algorithm as described in any one of claims 1 to 5.

8. A computer program product, comprising a computer program, characterized in that When the computer program is executed by a processor, the steps of the MIMO sparse array optimization method based on a genetic algorithm are implemented as described in any one of claims 1 to 5.

9. A radar, characterized in that: The MIMO sparse array optimization method based on genetic algorithm described in any one of claims 1 to 5 is used to perform MIMO sparse array optimization.

10. A vehicle, characterized in that: Comprising the radar as claimed in claim 9.

Images