A correlation-based array configuration optimization design method in multipath scenarios

Abstract

The application relates to a distributed radar array configuration optimization method under a multipath environment and belongs to the technical field of radar signal processing, in particular to an array configuration optimization method based on inter-array correlation. A double-target particle swarm optimization algorithm is adopted to optimize the array configuration of the distributed radar under the multipath environment, inter-array correlation is improved, grating lobes are suppressed, and the optimal array position which is less affected by the multipath is obtained. The application preferably solves the problem that the distributed radar is affected by the multipath and the signal processing performance is reduced under the multipath environment, and the optimal array position under the multipath environment can be calculated. The effectiveness of the optimization algorithm is verified through simulation experiments.

Description

Technical Field

[0001] This invention relates to a method for optimizing the configuration of distributed radar arrays in multipath scenarios, belonging to the field of radar signal processing technology, and particularly to an array configuration optimization method based on inter-array correlation. Background Technology

[0002] Distributed radar systems consist of multiple unit radars deployed separately. Through signal-level fusion processing, they form a large-aperture sparse array with a high-gain, extremely narrow beam, allowing for more advanced signal processing algorithms to detect and accurately locate various targets.

[0003] During target detection, the radar receives multipath echoes in addition to the target echo, which are also reflected or diffracted from the ground or other objects. The varying time delays of these multipath echoes reaching the receiver affect radar system performance, including changes in received echoes and deterioration in system performance. For distributed arrays, this also leads to decreased inter-array correlation. Considering the complexities of real-world transmission environments, multipath effects cannot be completely avoided.

[0004] Most existing literature focuses on the suppression and compensation of multipath effects. Qiu Wangsheng of the Air Force Early Warning Academy proposed a method using compensation factors to compensate for multipath signals. This method introduces compensation factors to account for the terrain environment near the radar, establishes a functional relationship between the compensation factors and the terrain environment, and uses the compensation factors to compensate for multipath signals. Cheng Jie of the University of Electronic Science and Technology of China proposed a method for detecting low-altitude moving targets in multipath environments. By adding frequency offset to the array elements, the frequency components of the transmitted signal are increased, thereby suppressing multipath, reducing the probability of echo signal cancellation, and improving the detection performance of low-altitude targets. Ma Ke of the Xi'an Institute of Electronic Engineering proposed a method for suppressing multipath effects in active protection radar. Using a data processing algorithm based on group selection, combined with traditional multipath suppression methods, outlier elevation angle values ​​affected by multipath effects are eliminated while the true values ​​are preserved, effectively suppressing the influence of multipath effects and improving radar measurement accuracy.

[0005] For distributed arrays, array configuration optimization is a crucial step. If the element spacing is inappropriate, the array's radiation pattern will develop similar radiation lobes outside the main lobe, known as grating lobes. When deploying arrays in multipath scenarios, both the array's radiation pattern performance and the impact of multipath on the array must be considered. However, currently available technologies lack a method to combat multipath from the perspective of array configuration optimization. Summary of the Invention

[0006] The technical problem solved by this invention is to overcome the shortcomings of existing technologies and propose a correlation-based array configuration optimization design method for multipath scenarios. The aim is to address the issue of multipath effects affecting radar in distributed arrays. This method introduces inter-array correlation as an evaluation index to assess the quality of radar received signals in multipath scenarios. A dual-target particle swarm optimization algorithm is employed, based on inter-array correlation and while also considering grating lobe suppression, to optimize the array configuration of distributed radar in multipath environments. This achieves optimal array placement with minimal multipath impact while maintaining array pattern performance.

[0007] The technical solution of this invention is:

[0008] A correlation-based array configuration optimization design method for multipath scenarios, comprising the following steps:

[0009] Step S1: Establish a distributed radar signal transmission model for a multipath scenario, including the transmitted signal model of the transmitting array and the echo signal model of the receiving array. The echo signal model of the receiving array includes a direct echo signal model and a multipath echo signal model. Then, set the parameters in the established distributed radar signal transmission model, including array position and array aperture.

[0010] Step S2: Based on the array position and array aperture obtained in step S1, calculate the composite radiation pattern of all arrays. The obtained composite radiation pattern includes the maximum grid lobe level.

[0011] Step S3: Based on the echo signals of the receiving arrays obtained in step S1, calculate the correlation between the arrays to obtain the correlation coefficient between the echo signals of the receiving arrays.

[0012] Step S4: The maximum grating lobe level obtained in step S2 and the correlation coefficient between the echo signal of the receiving array obtained in step S3 are optimized using the particle swarm optimization algorithm to find the result with the lowest maximum grating lobe level and the highest correlation coefficient as the optimization result.

[0013] Step S5: The optimal array position is the array position corresponding to the optimization result obtained in step S4.

[0014] In step S1, the specific method for establishing a distributed radar signal transmission model in a multipath scenario is as follows:

[0015] Step S11, set up a master array consisting of 1 transmit / receive co-located array and N r An antenna array consisting of several secondary receiving arrays, each of which is a planar phased array antenna. Taking the transmitting array as the reference array, the distance from the nth receiving array to the reference array is d. n n = 1, ..., N r +1, N rGiven the number of receiver arrays, assume the radar transmits a chirp signal with a pulse width of T. p The carrier frequency is f c If the frequency modulation slope is k, then the transmitted signal model s(t) of the transmitting array at time t is:

[0016]

[0017] The direct echo signal model for the nth receiving array is:

[0018] r n (t)=s(t-τ n (2)

[0019] Where, τ n This represents the echo delay of the nth receiving array;

[0020] Step S12: Since only the main array transmits signals and the main array area is large and the beam is narrow, the transmitted multipath is ignored and only the received multipath is considered. Based on this, a multipath echo signal model can be established.

[0021] Assuming multipath propagation originates from ground reflection points and is specular reflection, the multipath echo signal model for the nth receiving array is as follows:

[0022] h n (t)=pr n (t-τ mn (3)

[0023] Where, τ mn Let represent the multipath transmission delay of the nth receiving array, and p represent the specular reflection propagation loss, which can be considered a constant.

[0024] Then the echo signal model of the nth receiving array after adding the direct echo signal and the multipath echo signal is:

[0025] r mn (t)=r n (t)+h n (t) (4)

[0026] In step S2, the method for calculating the array composite radiation pattern based on the array position and array aperture obtained in step S1 is as follows:

[0027] Taking a one-dimensional receiver pattern as an example, the formula for calculating the array synthesized pattern is as follows:

[0028] P = F n *a n (5)

[0029] Where P represents the array composite pattern, F n This represents the radiation pattern of the nth receiver array, an Represents the steering vector of the nth receiver array:

[0030]

[0031]

[0032] Among them, A n Let θ represent the aperture of the nth receiving array, θ represent the scanning angle, and λ represent the signal wavelength.

[0033] Typically, in addition to a main lobe, an antenna pattern may have multiple grating lobes. The ratio of the maximum electromagnetic field intensity of the highest grating lobe to the maximum electromagnetic field intensity of the main lobe is the maximum grating lobe level, denoted as M. GL , expressed in decibels.

[0034] In step S3, the method for calculating the inter-array correlation based on the echo signal of the receiving array obtained in step S1 is as follows:

[0035] The correlation coefficients are calculated pairwise for the echo signals from the receiving array. The formula for calculating the correlation coefficients is as follows:

[0036]

[0037] Where r1(t) and r2(t) represent the echoes of the two arrays respectively, corr(·) represents the correlation between the two sets of data, Cov(·) represents the covariance between the two sets of data, E(·) represents the data variance, and D(·) represents the data variance.

[0038] The correlation coefficient between the echo signals of the receiving array is used as an evaluation index to measure the quality of the received signal. That is, the higher the correlation coefficient between the echo signals of the receiving array, the higher the quality of the received signal and the less affected it is by multipath propagation.

[0039] In step S4, the correlation coefficient between the maximum grating lobe level obtained in step S2 and the echo signal of the receiving array obtained in step S3 is optimized using the particle swarm optimization algorithm.

[0040] Step S4 further includes the following steps:

[0041] Step S41 involves optimizing using a dual-objective particle swarm optimization algorithm to find the result with the lowest maximum gate lobe level and the highest correlation coefficient. Particle swarm optimization (PSO) is an evolutionary computational technique that starts with random solutions and iteratively searches for the optimal solution. This algorithm has attracted attention due to its ease of implementation, high accuracy, and fast convergence, and has demonstrated its superiority in solving practical problems.

[0042] PSO is initialized as a swarm of random particles (random solutions), with multiple particles coexisting and cooperating to find the best solution. In each iteration, a particle updates itself by tracking two "extremes". The first is the optimal solution found by the particle itself, called the individual extreme value. The other extreme value is the optimal solution found by the entire population so far, which is the global extreme value.

[0043] A particle has only two attributes: velocity and position. Velocity represents how fast it moves, and position represents the direction of movement. In each iteration, the particle updates itself by tracking two optimal values ​​(pbest and gbest). After finding these two optimal values, the particle updates its velocity and position using the following formula.

[0044] v i =ω*v i +c1*rand()*(pbest i -x i )+c2*rand()*(gbest i -x i (9)

[0045] x i =x i +v i ,i=1,2,…,N (10)

[0046] Among them, v i Let ω represent the velocity of the i-th particle, ω represent the inertia factor (which can be updated with iterations or set to a constant), N represent the number of particles, and rand() represent a random number between 0 and 1. i pbest represents the current velocity of the i-th particle. i gbest represents the individual optimal value of the i-th particle. i Let c1 and c2 represent the global optimal value of the i-th particle, and c1 and c2 represent the learning factors.

[0047] For 1 master array and N r There are N auxiliary arrays in total. r +1 receiver arrays. Let the echo signal of the x-th receiver array be r. x (t), the echo signal of the y-th receiving array is r y If (t), then the optimization problem can be expressed as the following formula:

[0048] min{M GL} (11)

[0049] min{-corr(r x (t),r y (t))} (12)

[0050] stx,y∈(N r +1) (13)

[0051] Step S42: To prevent the algorithm from getting stuck in local optima and to make it easier to find the global optimum, mutation is introduced. Particles will mutate with a certain probability. During mutation, the particle will randomly jump from its current position to a nearby position, which helps it escape local optima and find the global optimum.

[0052] Beneficial effects

[0053] This invention proposes an array configuration optimization method based on inter-array correlation in a multipath environment. It has the following beneficial effects:

[0054] (1) In the method of the present invention, the correlation between the echo signals of the receiving array is calculated, and the echo quality is evaluated by the correlation coefficient between the echo signals of the receiving array. It can effectively measure the extent to which a distributed array is affected by multipath in complex multipath environments.

[0055] (2) In the method of this invention, a dual-objective particle swarm optimization algorithm is used for optimization, and the optimization result is the Pareto solution set. This optimizes the correlation between arrays while also suppressing grating lobes. It achieves the effect of reducing multipath while maintaining the array pattern performance.

[0056] (3) This invention effectively solves the problem of signal processing performance degradation of distributed radar under multipath environment. It can effectively reduce the impact of multipath on distributed radar and improve echo quality while maintaining array pattern performance. The optimization effect of the array configuration optimization method based on inter-array correlation under multipath environment is verified by simulation experiments. Attached Figure Description

[0057] Figure 1 This is a flowchart of the method of the present invention;

[0058] Figure 2 The flowchart of the particle swarm optimization algorithm used in this invention is shown below;

[0059] Figure 3 This is a schematic diagram of the array configuration and multipath propagation model;

[0060] Figure 4 To optimize the Pareto solution set;

[0061] Figure 5 The result is a weighted optimization graph;

[0062] Figure 6 To optimize the composite radiation pattern of the arrays before and after. Detailed Implementation

[0063] To make the objectives, technical solutions, and advantages of this invention clearer, the invention will be further described in detail below with reference to specific examples and the accompanying drawings.

[0064] The array configuration optimization method based on inter-array correlation in multipath environments proposed in this invention, such as... Figure 1 As shown, the steps are as follows:

[0065] Step S1: Establish a distributed radar signal transmission model for a multipath scenario, including the transmitted signal model of the transmitting array and the echo signal model of the receiving array. The echo signal model of the receiving array includes a direct echo signal model and a multipath echo signal model. Then, set the parameters in the established distributed radar signal transmission model, including array position and array aperture.

[0066] Step S2: Based on the array position and array aperture obtained in step S1, calculate the composite radiation pattern of all arrays. The obtained composite radiation pattern includes the maximum grid lobe level.

[0067] Step S3: Based on the echo signals of the receiving arrays obtained in step S1, calculate the correlation between the arrays to obtain the correlation coefficient between the echo signals of the receiving arrays.

[0068] Step S4: The maximum grating lobe level obtained in step S2 and the correlation coefficient between the echo signal of the receiving array obtained in step S3 are optimized using the particle swarm optimization algorithm to find the result with the lowest maximum grating lobe level and the highest correlation coefficient as the optimization result.

[0069] Step S5: The optimal array position is the array position corresponding to the optimization result obtained in step S4.

[0070] In step S1, the method for establishing a distributed radar signal transmission model under multipath environment is as follows:

[0071] Step S11, set up a master array consisting of 1 transmit / receive co-located array and N r An antenna array consisting of several secondary receiving arrays, each of which is a planar phased array antenna. Taking the transmitting array as the reference array, the distance from the nth receiving array to the reference array is d. n n = 1, ..., N r +1, N r Given the number of receiver arrays, assume the radar transmits a chirp signal with a pulse width of T. p The carrier frequency is f c If the frequency modulation slope is k, then the transmitted signal model s(t) of the transmitting array at time t is:

[0072]

[0073] The direct echo signal model for the nth receiving array is:

[0074] r n (t)=s(t-τ n (15)

[0075] Where, τ n This represents the echo delay of the nth receiving array.

[0076] Step S12: Since only the main array transmits signals and the main array area is large and the beam is narrow, the transmitted multipath is ignored and only the received multipath is considered. Based on this, a multipath echo signal model can be established.

[0077] Assuming multipath propagation originates from ground reflection points and is specular reflection, the multipath echo signal model for the nth receiving array is as follows:

[0078] h n (t)=pr n (t-τ mn (16)

[0079] Where, τ mn Let represent the multipath transmission delay of the nth receiving array, and p represent the specular reflection propagation loss, which can be considered a constant.

[0080] Then the echo signal model of the nth receiving array after adding the direct echo signal and the multipath echo signal is:

[0081] r mn (t)=r n (t)+h n (t) (17)

[0082] In step S2, the method for calculating the array composite radiation pattern based on the array position and array aperture obtained in step S1 is as follows:

[0083] Taking a one-dimensional receiver pattern as an example, the formula for calculating the array synthesized pattern is as follows:

[0084] P = F n *a n (18)

[0085] Where P represents the array composite pattern, F n This represents the radiation pattern of the nth receiver array, a n Represents the steering vector of the nth receiver array:

[0086]

[0087]

[0088] Among them, A nLet θ represent the aperture of the nth receiving array, θ represent the scanning angle, and λ represent the signal wavelength.

[0089] Typically, in addition to a main lobe, an antenna pattern may have multiple grating lobes. The ratio of the maximum electromagnetic field intensity of the highest grating lobe to the maximum electromagnetic field intensity of the main lobe is the maximum grating lobe level, denoted as M. GL , expressed in decibels.

[0090] In step S3, the method for calculating the inter-array correlation based on the echo signal of the receiving array obtained in step S1 is as follows:

[0091] The correlation coefficients are calculated pairwise for the echo signals from the receiving array. The formula for calculating the correlation coefficients is as follows:

[0092]

[0093] Where r1(t) and r2(t) represent the echoes of the two arrays respectively, corr(·) represents the correlation between the two sets of data, Cov(·) represents the covariance between the two sets of data, E(·) represents the data variance, and D(·) represents the data variance.

[0094] The correlation coefficient between the echo signals of the receiving array is used as an evaluation index to measure the quality of the received signal. That is, the higher the correlation coefficient between the echo signals of the receiving array, the higher the quality of the received signal and the less affected it is by multipath propagation.

[0095] In step S4, the correlation coefficient between the maximum grating lobe level obtained in step S2 and the echo signal of the receiving array obtained in step S3 is optimized using the particle swarm optimization algorithm.

[0096] Step S4 further includes the following steps:

[0097] Step S41 involves using a dual-objective particle swarm optimization algorithm to find the result with the lowest maximum gate lobe level and the highest correlation coefficient. For example... Figure 2 As shown, Particle Swarm Optimization (PSO) is an evolutionary computational technique that starts with random solutions and iteratively searches for the optimal solution. This algorithm has attracted attention due to its ease of implementation, high accuracy, and fast convergence, and has demonstrated its superiority in solving practical problems.

[0098] PSO is initialized as a swarm of random particles (random solutions), with multiple particles coexisting and cooperating to find the best solution. In each iteration, a particle updates itself by tracking two "extremes". The first is the optimal solution found by the particle itself, called the individual extreme value. The other extreme value is the optimal solution found by the entire population so far, which is the global extreme value.

[0099] A particle has only two attributes: velocity and position. Velocity represents how fast it moves, and position represents the direction of movement. In each iteration, the particle updates itself by tracking two optimal values ​​(pbest and gbest). After finding these two optimal values, the particle updates its velocity and position using the following formula.

[0100] v i =ω*v i +c1*rand()*(pbest i -x i )+c2*rand()*(gbest i -x i ) (twenty two)

[0101] x i =x i +v i ,i=1,2,…,N (23)

[0102] Among them, v i Let ω represent the velocity of the i-th particle, ω represent the inertia factor (which can be updated with iterations or set to a constant), N represent the number of particles, and rand() represent a random number between 0 and 1. i pbest represents the current velocity of the i-th particle. i gbest represents the individual optimal value of the i-th particle. i Let c1 and c2 represent the global optimal value of the i-th particle, and c1 and c2 represent the learning factors.

[0103] For 1 master array and N r There are N auxiliary arrays in total. r +1 receiver arrays. Let the echo signal of the x-th receiver array be r. x (t), the echo signal of the y-th receiving array is r y If (t), then the optimization problem can be expressed as the following formula:

[0104] min{M GL} (twenty four)

[0105] min{-corr(r x (t),r y (t))} (25)

[0106] stx,y∈(N r +1) (26)

[0107] Step S42: To prevent the algorithm from getting stuck in local optima and to make it easier to find the global optimum, mutation is introduced. Particles will mutate with a certain probability. During mutation, the particle will randomly jump from its current position to a nearby position, which helps it escape local optima and find the global optimum.

[0108] The following is an example of array configuration optimization in a multipath environment using the above method.

[0109] Example

[0110] Distributed radar signal transmission model in multipath environment, such as Figure 3 As shown. Assume the distributed radar system is S-band, with one main array (transmit and receive) and four auxiliary arrays (receive only). The arrays are arranged with a 100m baseline, and the auxiliary arrays at both ends are fixed. Array configuration optimization is performed, and the deployment positions of the remaining auxiliary arrays are determined based on the optimization results.

[0111] A distributed radar array is deployed in open terrain, with the target located 100 km away. The target echo undergoes multipath reflection at the ground in front of the array. Due to the complex terrain of the multipath reflection zone, there are numerous multipath echoes with inconsistent intensities, and the multipath intensity varies across the baseline. Since the main array has a large area and a narrow beam, it is assumed to be unaffected by multipath. Therefore, to simplify calculations, the main array can be used as a reference array, and only the correlation between the main and auxiliary arrays needs to be considered when calculating the inter-array correlation. Considering that there are only two auxiliary arrays to be optimized, the average correlation between the main array and the two auxiliary arrays is used to represent the main-auxiliary array correlation.

[0112] The radar waveform parameters and source parameters are shown in Table 1 and Table 2, respectively.

[0113] Table 1 Radar waveform parameters

[0114]

[0115] Table 2 Source Parameters

[0116]

[0117] The iterative output of the dual-objective particle swarm optimization algorithm is as follows: Figure 4 As shown. Under the current scenario parameter constraints, the correlation between the primary and secondary components and the maximum grating lobe level cannot simultaneously reach the optimal state, meaning that an absolutely optimal solution does not exist. Therefore, considering both optimization objectives, a Pareto optimal solution is introduced, resulting in a Pareto solution set. Each solution in the Pareto solution set corresponds to a set of optimal array positions.

[0118] We obtain the weighted objective function s by weighting the sum of the two optimization objectives. t :

[0119] st =k1*M SL +k2*(-cor) (27)

[0120] Where k1 and k2 represent weights, M SL represents the maximum grating lobe level, and cor represents the correlation between the primary and secondary arrays.

[0121] Figure 5 The algorithm shows the decreasing trend of the weighted objective function as the algorithm iterates. It can be seen that the employed dual-objective particle swarm optimization algorithm converges and yields an optimal result.

[0122] The synthesized radiation pattern of the array under uniform array arrangement before optimization and the synthesized radiation pattern of the array after optimization are shown below. Figure 6 As shown in the figure, the maximum grid lobe level is significantly reduced after optimization.

[0123] The specific values ​​of the optimization results are shown in Table 3. It can be seen that through the optimization algorithm, the maximum grid lobe level is reduced by about 1.3dB and the correlation between the main and auxiliary arrays is improved by about 0.3, which effectively reduces the optimization target value and obtains good optimization results.

[0124] Table 3 Optimization Results

[0125]

[0126] This invention may have many other embodiments, and the above are merely preferred embodiments of the invention and are not intended to limit the scope of protection of the invention. Any modifications, equivalent substitutions, improvements, etc., made within the spirit and principles of this invention should be included within the scope of protection of this invention.

Claims

1. A correlation-based array configuration optimization design method in a multipath scenario, characterized in that The steps of the method include: Step S1, a distributed radar signal transmission model under a multipath scene is established, and an array position and an array aperture of the established distributed radar signal transmission model are obtained; the distributed radar signal includes a transmitting signal of a transmitting array and a return signal of a receiving array, wherein the return signal of the receiving array includes a direct return signal and a multipath return signal; Step S2, according to the array position and the array aperture obtained in step S1, a synthesized directivity diagram of all arrays is calculated, and a maximum grating lobe level is obtained according to the synthesized directivity diagram; Step S3, according to the return signal of the receiving array in step S1, an inter-array correlation is calculated, and a correlation coefficient between the return signals of the receiving array is obtained; Step S4, the maximum grating lobe level obtained in step S2 and the correlation coefficient between the return signals of the receiving array obtained in step S3 are used to find a result with the lowest maximum grating lobe level and the highest correlation coefficient as an optimization result; Step S5, the array position corresponding to the optimization result obtained in step S4 is used as an optimal array position, and an array configuration optimization design under a multipath scene based on correlation is completed; In step S4, a double-target particle swarm optimization algorithm is used for optimization, a result with the lowest maximum grating lobe level and the highest correlation coefficient is found, and a group of random particles is initialized; multiple particles coexist and cooperate in optimization; in each iteration, a particle is updated by tracking two “extreme values”; the first extreme value is an optimal solution found by the particle, and the solution is an individual extreme value; the other extreme value is an optimal solution found by the entire population, and the solution is a global extreme value; In each iteration, a particle is updated by tracking two optimal values; after the two optimal values are found, the particle is updated by using the following formula to update a speed and a position; in, Indicates the first The speed of each particle This represents the inertia factor, which is updated with each iteration or set to a constant. Indicates the number of particles. Represents a random number between 0 and 1. Indicates the first The current velocity of each particle Indicates the first The individual optimal value of each particle. Indicates the first The global optimal value for each particle. , This represents the learning factor.

2. The array configuration optimization design method under a multipath scene based on correlation according to claim 1, characterized in that: In step S1, a specific method for establishing the distributed radar signal transmission model under the multipath scene is as follows: Step S11, set an antenna array composed of 1 transceiver co-located main array and 1 receiving auxiliary array, each antenna in the array is a flat panel phased array antenna, taking the transmitting array as the reference array, the distance from the first receiving array to the reference array is , , +1, the number of receiving arrays, assuming that the radar transmits a chirp signal, the pulse width is , the carrier frequency is , and the frequency modulation slope is , then the transmission signal model of the transmitting array at time is: The direct echo signal model for the mth receiving array is then given by: ​ wherein, represents the time delay of the nth receiving array echo. Step S12, a multipath return signal model is established; Let the multipath be generated by ground reflection points and be specular reflection, the first received array multipath echo signal model is: wherein, represents the th received array multipath transmission delay, represents the specular reflection propagation loss; The echo signal model for the direct echo signal plus the multipath echo signal of the mth receiving array is: ​ 3. The array configuration optimization design method under a multipath scene based on correlation according to claim 2, characterized in that: In step S2, according to the array position and the array aperture obtained in step S1, the array synthesized directivity diagram is calculated as follows: wherein, represents an array synthesis pattern, represents the th receive array pattern, represents the th receive array steering vector: wherein, represents the number of receiving array apertures, represents the number of receiving array apertures, represents the scan angle, represents the signal wavelength.

4. The array configuration optimization design method under a multipath scene based on correlation according to claim 3, characterized in that: The maximum grating lobe level in step S2 is the ratio of the maximum electromagnetic field intensity of the highest grating lobe to the maximum electromagnetic field intensity of the main lobe according to the synthesis direction diagram, denoted as in decibels.

5. The array configuration optimization design method under a multipath scene based on correlation according to claim 4, characterized in that: In step S3, according to the return signal of the receiving array obtained in step S1, a method for calculating the inter-array correlation is as follows: A correlation coefficient is calculated for the return signals of the receiving array two by two, and a correlation coefficient calculation formula is as follows: wherein, and denote the echo of two arrays, respectively, denotes the correlation of two sets of data, denotes the covariance of two sets of data, denotes the variance of data, denotes the variance of data.

6. The array configuration optimization design method under a multipath scene based on correlation according to claim 5, characterized in that: The correlation coefficient between the return signals of the receiving array is used as an evaluation index to measure the advantages and disadvantages of the received signals; the higher the correlation coefficient between the return signals of the receiving array, the higher the quality of the received signals, and the less the received signals are affected by multipath.

7. The method of claim 6, wherein the method is characterized by: In the algorithm, mutation is introduced, and the particles will mutate with a set probability. When mutating, the particles will randomly jump from the current position to the nearby position, which helps to jump out of the local optimal position and find the global optimal solution.

Images