Multi-view MIMO radar correlated imaging radiation field design method based on image entropy

Abstract

The invention discloses a multi-view MIMO radar correlated imaging radiation field design method based on image entropy, which considers the fluctuation of the scattering intensity of each resolution unit of a target in an imaging scene under a multi-view condition, carries out statistical modeling on the scattering intensity of each resolution unit of the target, and starts from the whole process of transmitting, receiving and processing a signal. By constraining the condition number of a radiation field Gram matrix and the image entropy of imaging and through continuous iterative optimization, an array type of an MIMO radar transmitting waveform and a transmitting array which have a good suppression effect on target grid energy dissipation and are stable in imaging is found, and the correlation between a radiation field reference matrix and signal echoes is improved. According to the method, radar correlated imaging can be carried out on the target with fluctuating scattering intensity of each resolution unit by using the optimized emission signal waveform and the array of the emission array, the randomness of a radiation field is enhanced, the energy dissipation degree of a target grid is reduced, the imaging error is reduced, and a better radar imaging result can be obtained.

Description

Technical Field

[0001] The invention belongs to the field of radar correlation imaging signal processing, takes into account the fluctuation of scattering intensity of each resolution unit of a target, and specifically relates to a multi-view MIMO radar correlation imaging radiation field design method based on image entropy. Background Art

[0002] Radar has the ability to work all day, all weather and over long distances, and has received much attention and extensive research. In recent years, by drawing on the principle of optical ghost imaging, radar correlation imaging has developed into a new high-resolution staring imaging technology that does not rely on the relative motion between the radar and the target, providing a new perspective and solution to the bottleneck problems of imaging stationary, quasi-stationary targets and non-cooperative targets in existing radar imaging technology.

[0003] Radar correlation imaging divides the imaging plane into grids, and forms a two-dimensional random radiation field in space and time in the imaging area by emitting randomly modulated signals. The distribution of the target is obtained by correlating the echo with the reference signal of each grid. The two-dimensional random radiation field in space and time provides sufficient information for the resolution and imaging of the target, and determines the quality of correlation imaging. Constructing a two-dimensional random radiation field in space and time is a prerequisite for radar correlation imaging. Many scholars have studied the problem of weak randomness of the radiation field in radar correlation imaging and achieved good results. Existing studies have analyzed the problem of random characterization of the radiation field, such as using effective rank to evaluate the randomness of the radiation field, mainly designing the waveform of the transmitted signal and the formation of the radar antenna array; there are also waveform design based on condition number, such as waveform and transmission formation design based on minimizing the difference between the Gram matrix of the radiation field and the unit matrix.

[0004] The performance, viewing angle, polarization mode, etc. of the target itself will cause changes in the overall RCS of the point target, and will also cause changes in the scattering intensity of each resolution unit of the extended target. After radar imaging, the comprehensive value of each pixel point of the radar image is obtained. When considering the fluctuations in the scattering intensity of each resolution unit of the extended target, it is found that the reconstructed image is relatively chaotic, indicating that the correlation between the reference matrix and the scattered echo is degraded and the target grid energy is dissipated. Previous studies did not consider the scattering characteristics of each resolution unit of the target when designing the radiation field, so the use of the optimized random radiation field for imaging cannot achieve a good reconstruction effect.

[0005] The present invention mainly aims at the situation that the MIMO radar has poor correlation imaging effect on the target with fluctuating scattering intensity of each resolution unit and the correlation between the reference matrix and the echo is degraded, and proposes a multi-view MIMO radar correlation imaging radiation field design method based on image entropy. Starting from the overall process of signal transmission, reception and processing, the scattering intensity of each resolution unit of the target is statistically modeled, and a MIMO radar transmission waveform and transmission array formation with good suppression effect on target grid energy dissipation and relatively stable imaging are sought, thereby improving the correlation between the radiation field reference matrix and the signal echo, and improving the image effect. Summary of the invention

[0006] Purpose of the invention: The present invention provides a multi-view MIMO radar correlation imaging radiation field design method based on image entropy to solve the problem that the scattering characteristics of the actual target are not considered in the traditional radar correlation imaging radiation field design, resulting in poor imaging effect for targets with fluctuating scattering intensities of each resolution unit.

[0007] Invention content: The multi-view MIMO radar correlation imaging radiation field design method based on image entropy of the present invention comprises the following steps:

[0008] (1) Set up the MIMO radar correlation imaging scenario to generate the transmit signal waveform and transmit array element position.

[0009] (2) According to the waveform of the transmitted signal, the position of the transmitting and receiving array elements, and the position of the imaging unit, N reference matrices A are obtained. n , concatenate the N reference matrices in the time dimension to obtain the expanded-dimensional reference matrix A = [A1; A2; ···; A n ]; statistical modeling is performed on each resolution unit of the target. After the transmitted signal is scattered by each resolution unit, it is sampled at the receiving end for J time to obtain N echo vectors y = [y1; y2; ···; y n ], and the expanded-dimensional echo vector y is obtained after splicing;

[0010] (3) Based on the sparse reconstruction multi-view MIMO radar correlation imaging algorithm, the expanded-dimensional reference matrix and the expanded-dimensional echo are jointly processed to obtain the scattering intensity value of each resolution unit of the imaging scene;

[0011] (4) Set the maximum number of iterations Z max , taking the radiation field Gram matrix condition number and the image entropy obtained from the scattering intensity value of each resolution unit of the imaging scene as the objective function, the waveform of the transmitting signal and the formation of the transmitting array are iteratively optimized by constraining the radiation field Gram matrix condition number and image entropy;

[0012] (5) If the iteration times are not reached, return to step (2); if the iteration times are reached, output the optimal signal waveform and the array configuration of the transmitting array.

[0013] Furthermore, the implementation process of the scenario setting in step (1) is as follows:

[0014] The number of transmitting array elements of the MIMO radar is M, the number of receiving array elements is N, the transmitting array is a random linear array, the receiving array is a uniform linear array, and the transmitted signal uses a random frequency hopping signal; the imaging plane is divided into L = x num *y num imaging units, where x num represents the number of horizontal imaging units, y num represents the number of vertical imaging units, and the number of time samplings is J.

[0015] Furthermore, the waveform of the transmitted signal in step (1) is:

[0016]

[0017] where the waveform of the random frequency hopping signal transmitted by the m-th transmitting array element is S m (t), which consists of Q sub-pulses, q represents the q-th sub-pulse, f c represents the carrier frequency, Δt and Δf represent the pulse duration and the minimum frequency interval, G is a positive integer, c m,q is the frequency hopping code corresponding to each sub-pulse, c m,q ∈{0, 1, … G - 1}; u(t) represents a gate function that takes a value of 1 when 0 < t < Δt and 0 otherwise.

[0018] Furthermore, the positions of the transmitting array elements in step (1) are:

[0019] The position p1 of the first array element in the transmitting array is 0, the position p M of the last array element is F, and the positions p m of the other array elements except the first and last ones ∈{1, 2, ··· F - 1}, m = 2, 3, ··· M - 1; F is a positive integer;

[0020] The position P of the transmitting array element to be optimized is represented by the following formula:

[0021] P = {p m |p m ∈{1, 2, … F - 1}, if i ≠ j then p i ≠ p j , m, i, j = 2, 3, … M - 1} (3).

[0022] Furthermore, the implementation process of statistical modeling of each resolution unit of the target in step (2) is as follows:

[0023] The scattering intensity σ of each resolution unit is statistically modeled using a complex Gaussian model, σ = A0e jφ , σ obeys complex Gaussian distribution, the amplitude A0 of σ obeys Rayleigh distribution, and the phase φ of σ obeys uniform distribution:

[0024]

[0025] Among them, f(A0) is the probability density of amplitude, f(φ) is the probability density of phase, A n (r l ,t) is the reference signal, y n (t) is the echo signal, S m (t) represents the transmission signal of the mth transmitting array element, c is the propagation speed of electromagnetic waves, is the path delay from the mth transmitting array element through the lth imaging unit to the nth receiving array element; is the scattering intensity of the lth imaging unit at the viewing angle from the mth transmitting element to the nth receiving element; w n (t) is the noise of the nth receiving channel.

[0026] Furthermore, the step (3) is implemented by the following formula:

[0027]

[0028] Among them, ||·||1 represents the l1 norm of the vector, ||·||2 represents the l2 norm of the vector, and μ is the regularization parameter, which is used to control the balance between the l1 regularization term and the data fitting term. By minimizing the objective function, the scattering vector of the imaging target that meets the sparse condition is obtained.

[0029] Furthermore, step (4) is implemented by the following formula:

[0030]

[0031] Among them, A is the expanded radiation field reference matrix, A H represents the conjugate transpose of A, cond(·) represents the condition number, h(v) represents the image entropy, and v l represents the estimated value of the scattering intensity of each imaging unit, ||·||1 represents the l1 norm of the vector, ξ represents the optimization weight, C is the frequency hopping coding matrix, and P is the non-repeated transmitting array element position coding set.

[0032] Furthermore, the iterative optimization solution described in step (4) is implemented by the gray wolf optimization algorithm, and the process is as follows:

[0033] The GWO algorithm models the encirclement behavior using the following equations:

[0034] D=|C'·X p (z)-X(z)| (10)

[0035] X(z+1)=X p (z)-E·D (11)

[0036] E=2a·ε1-a (12)

[0037] C'=2·ε2 (13)

[0038] Among them, z represents the current iteration number, X p (z) is the position vector of the prey, X(z) represents the position vector of the gray wolf, a is the convergence factor, ε1 and ε2 are random vectors between [0,1];

[0039] The GWO algorithm simulates hunting behavior using the following equation:

[0040] D α =|C′1·X α -X|D β =|C′2·X β -X|D δ =|C′3·X δ -X| (14)

[0041] X1=X α -E1·D α X2=X β -E2·D β X3=X δ -E3·D δ (15)

[0042]

[0043] Each set of objective functions obtained by different transmitting waveforms and transmitting array element positions is called an individual, and α, β, and δ wolves are the three individuals with the best, second best, and third best performance, D α , D β , D δ That is, the gap between the remaining individuals and the top three individuals, X α , X β , X δ That is, the transmission waveforms and array element positions of the three individuals with the top three performances, C'1, C'2, and C'3 are vectors randomly generated in the process, X is the transmission signal waveform and transmission array formation of the individual, and X(z+1) is the transmission signal waveform and transmission array formation corresponding to the individual after update.

[0044] Beneficial effects: Compared with the prior art, the present invention has the following beneficial effects:

[0045] The present invention adopts a MIMO radar correlation imaging system, takes into account the scattering characteristics of actual targets, and aims at the problems of poor imaging effect when imaging targets with fluctuating scattering intensities of actual resolution units and degradation of correlation between reference matrix and echo. Starting from the whole process of signal transmission, reception and processing, the Gram matrix condition number of the radiation field expansion reference matrix and the image entropy of imaging are constrained to design the transmission signal waveform of the radiation field and the formation of the transmission array. Experimental results show that the transmission signal waveform and the transmission formation obtained by iterative optimization using the method are used for imaging, which enhances the randomness of the radiation field, improves the imaging performance, reduces the energy dissipation degree of the target grid, reduces the imaging error, and can obtain better radar imaging results. BRIEF DESCRIPTION OF THE DRAWINGS

[0046] Figure 1 is a flow chart of the present invention;

[0047] Figure 2 The waveform of the random frequency hopping signal used in the present invention;

[0048] Figure 3 is the position of the transmitting array element in the present invention;

[0049] Figure 4 This is a schematic diagram of the MIMO radar correlation imaging principle of the present invention;

[0050] Figure 5 is the ideal imaging scene diagram;

[0051] Figure 6 The imaging results are obtained when imaging the same target scene using different transmission waveforms and transmission element positions; wherein, (a) is the result obtained by imaging using an unoptimized transmission signal random waveform and a transmission element random position; (b) is the result obtained by imaging using a transmission signal waveform and a transmission array formation optimized with maximizing the effective rank as the objective function; (c) is the result obtained by imaging using a transmission signal waveform and a transmission array formation optimized with minimizing the difference between the radiation field Gram matrix and the unit matrix as the objective function; (d) is the result obtained by imaging using a transmission signal waveform and a transmission element formation optimized with minimizing the condition number of the radiation field Gram matrix as the objective function; (e) is the result obtained by imaging using a transmission signal waveform optimized with uniform transmission element positions and with minimizing the condition number of the radiation field Gram matrix and the entropy of the imaging image as the objective function; (f) is the result obtained by imaging using a transmission signal waveform and a transmission array formation optimized with minimizing the condition number of the radiation field Gram matrix and the entropy of the imaging image as the objective function. Detailed implementation manners

[0052] The present invention will be further described below with reference to the accompanying drawings.

[0053] The present invention provides a multi-view MIMO radar correlation imaging radiation field design method based on image entropy, including the following steps:

[0054] Step 1: Set the scene for radar correlation imaging, including setting the number and type of transmitting and receiving array elements, the division method of the imaging grid, the type of transmitted signal, the number of time samplings, etc.

[0055] The number of transmitting array elements of the MIMO radar is M, the number of receiving array elements is N, the transmitting array is a random linear array, and the receiving array is a uniform linear array. A group of random frequency hopping signals are transmitted at the transmitting antenna end. The imaging plane is divided into L = x num *y num imaging units. It is assumed that the target scattering center is located at the center of the imaging unit, where x num represents the number of horizontal imaging units, and x num represents the number of vertical imaging units. The number of time samplings is J.

[0056] The waveform of the random frequency hopping signal is specifically as Figure 2 shown.

[0057]

[0058] Among them, the waveform of the random frequency hopping signal transmitted by the m-th transmitting array element is S m (t), which consists of Q sub-pulses. q represents the q-th sub-pulse, f c represents the carrier frequency, Δt and Δf represent the pulse duration and the minimum frequency interval, G is a positive integer, c m,q is the frequency hopping code corresponding to each sub-pulse, and c m,q ∈{0, 1, … G - 1}. u(t) represents a gate function that takes the value of 1 within 0 < t < Δt and 0 otherwise.

[0059] The position of the transmitting array element to be optimized is expressed by the formula:

[0060] P = {p m |p m ∈{1, 2, … F - 1}, if i ≠ j then p i ≠ p j , m, i, j = 2, 3, … M - 1} (3)

[0061] The formation of the transmitting array is as Figure 3 shown, where the solid circles are the positions of the transmitting array, and the hollow circles are the vacant positions between the array elements. λ is the signal wavelength, and the array aperture is The position of the first array element is p1 = 0, and the position of the last array element is p M =F, the position p of other array elements to be optimized except the first and last array elements m ∈{1,2,···F-1},m=2,3,···M-1, and the position of each array element is not repeated, F is a positive integer.

[0062] Step 2: The schematic diagram of radar correlation imaging is as follows Figure 4 Generate the waveform of the transmitting signal and the position of the transmitting array element, and obtain the delay of each path corresponding to the L imaging units according to the waveform of the transmitting signal, the position of the transmitting and receiving array elements, and the position of the imaging unit, and then obtain the reference matrix corresponding to each receiving array element, and then obtain N reference matrices [A n , concatenate the N reference matrices in the time dimension to obtain the expanded-dimensional reference matrix A = [A1; A2; ···; A n ]. Statistical modeling is performed on each resolution unit of the target. After the transmitted signal is scattered by each resolution unit, it is sampled at the receiving end for J time to obtain N echo vectors y n , after splicing, we get the expanded-dimensional echo vector y=[y1; y2; ···; y n ]. The echo signal can be regarded as a linear combination of the waveforms reflected by all the resolution units to all the transmitting array elements.

[0063] The scattering intensity σ of each resolution unit is statistically modeled using a complex Gaussian model, σ = A0e jφ , σ obeys complex Gaussian distribution, the amplitude A0 of σ obeys Rayleigh distribution, and the phase φ of σ obeys uniform distribution. The probability density expression of amplitude f(A0) is shown in formula (3), and the probability density of phase f(φ) is shown in formula (4):

[0064]

[0065] Among them, A n (r l ,t) is the reference signal, r l is the position vector of the center of the lth imaging unit in the imaging scene, R m is the position vector of the mth transmitting array element, R n is the position vector of the nth receiving array element, y n (t) is the echo signal, S m (t) represents the transmission signal of the mth transmitting array element, c is the propagation speed of electromagnetic waves, is the path delay from the mth transmitting array element through the lth imaging unit to the nth receiving array element; is the scattering intensity value of the lth imaging unit at the transmitting and receiving angle of view from the mth transmitting array element to the nth receiving array element; w n(t) is the noise of the nth receiving channel.

[0066] Step 3: Based on the sparse reconstruction multi-view MIMO radar correlation imaging algorithm, the expanded-dimensional reference matrix and the expanded-dimensional echo are jointly processed to obtain the scattering intensity value of each resolution unit of the imaging scene.

[0067]

[0068] Among them, ||·||1 represents the l1 norm of the vector, ||·||2 represents the l2 norm of the vector, and μ is the regularization parameter, which is used to control the balance between the l1 regularization term and the data fitting term. By minimizing this objective function, the scattering vector of the imaging target that meets the sparse condition is obtained.

[0069] Step 4: Set the maximum number of iterations Z max , taking the radiation field Gram matrix condition number and the image entropy obtained from the scattering intensity value of each resolution unit of the imaging scene as the objective function, the waveform of the transmission signal and the formation of the transmission array are iteratively updated by constraining the radiation field Gram matrix condition number and image entropy. The transmission signal waveform and the transmission array formation are optimized, as shown in formula (9):

[0070]

[0071] Among them, A is the expanded reference matrix, A H represents the conjugate transpose of A, cond(·) represents the condition number, h(v) represents the image entropy, and v l represents the estimated value of the scattering intensity of each imaging unit, ‖·‖1 represents the l1 norm of the vector, ξ represents the optimization weight, C is the frequency hopping coding matrix, and P is the non-repeating transmitting array element position coding set. The waveform design is performed by optimizing the frequency hopping coding matrix of the random frequency hopping signal, and the formation design is performed by optimizing the position of the transmitting array element. The condition number of the radiation field Gram matrix and the image entropy of the imaging are constrained to reduce the morbidity of the reference matrix, improve the randomness of the radiation field and the imaging effect, and reduce the energy dissipation of the target grid.

[0072] The present invention uses the Grey Wolf Optimizer (GWO) algorithm to perform iterative optimization. The GWO algorithm uses the following equation to simulate the encirclement behavior:

[0073] D=|C'·X p (z)-X(z)| (10)

[0074] X(z+1)=X p (z)-E·D (11)

[0075] E=2a·ε1-a (12)

[0076] C'=2·ε2 (13)

[0077] Among them, z represents the current iteration number, X p (z) is the position vector of the prey, X(z) represents the position vector of the gray wolf, a is the convergence factor, and ε1 and ε2 are random vectors between [0,1].

[0078] The GWO algorithm simulates hunting behavior using the following equation:

[0079] D α =|C′1·X α -X|D β =|C′2·X β -X|D δ =|C′3·X δ -X| (14)

[0080] X1=X α -E1·D α X2=X β -E2·D β X3=X δ -E3·D δ (15)

[0081]

[0082] The optimal solution, the second optimal solution, and the third optimal solution are called α, β, and δ wolves. α , D β , D δ represents the distance between α, β, δ wolves and other individuals, X α , X β , X δ is the current position vector of α, β, δ wolves, C'1, C'2, C'3 are random vectors obtained by equation (12), X is the current position of the gray wolf, and X(z+1) is the updated position of the gray wolf.

[0083] Each set of objective functions obtained by different transmitting waveforms and transmitting array element positions in this problem is called an individual, and α, β, and δ wolves are the three individuals with the best, second best, and third best performance, respectively. α , D β , D δ That is, the gap between the remaining individuals and the top three individuals, X α , X β , X δThat is, the transmission waveforms and array element positions of the three individuals with the top three performances, C'1, C'2, and C'3 are vectors randomly generated in the process, X is the transmission signal waveform and transmission array formation of the individual, and X(z+1) is the transmission signal waveform and transmission array formation corresponding to the individual after update.

[0084] Step 5: If the number of iterations has not been reached, return to step 2. If the number of iterations has been reached, output the optimal signal waveform and the formation of the transmitting array.

[0085] The entire iterative optimization process is as follows: 1) Initialize and generate random individuals, including the transmitted signal waveform encoding and the array element position encoding; 2) Calculate the radiation field reference matrix and echo according to the transmitted signal waveform and the array element position, and obtain the scattering intensity value of each resolution unit through sparse reconstruction, calculate the fitness of all individuals, i.e., the objective function, and record the positions of α, β, and δ; 3) Generate a, E, and C', update the positions of X1, X2, X3, and X(z+1) according to equations (14) and (15), and determine whether the array element position is repeated. If it is repeated, re-update the individual position; 4) Calculate the reference matrix and echo according to the updated individual and obtain the scattering intensity value of each resolution unit, calculate the fitness of all individuals after the update, and update the positions of α, β, and δ; 5) When the number of iterations is less than the maximum number of iterations, repeat 3) and 4). If the number of iterations reaches the maximum number of iterations, output the optimal individual, including the optimal waveform encoding and the transmitting array element position encoding.

[0086] Figure 5 When conducting simulation experiments, the present invention sets the scattering intensity of each resolution unit of the target to satisfy the complex Gaussian distribution with the transmission and reception viewing angle, with a mean of 10 dBsm and a variance of 8 dBsm as the ideal imaging scene diagram; Figure 6The imaging results are obtained when imaging the same target scene using different transmission waveforms and transmission element positions, wherein (a) is the result obtained by imaging using an unoptimized transmission signal random waveform and a transmission element random position; (b) is the result obtained by imaging using a transmission signal waveform and a transmission array formation optimized with maximizing the effective rank as the objective function; (c) is the result obtained by imaging using a transmission signal waveform and a transmission array formation optimized with minimizing the difference between the radiation field Gram matrix and the unit matrix as the objective function; (d) is the result obtained by imaging using a transmission signal waveform and a transmission element formation optimized with minimizing the condition number of the radiation field Gram matrix as the objective function; (e) is the result obtained by imaging using a transmission signal waveform optimized with uniform transmission element positions and with minimizing the condition number of the radiation field Gram matrix and the entropy of the imaging image as the objective function; (f) is the result obtained by imaging using a transmission signal waveform and a transmission array formation optimized with minimizing the condition number of the radiation field Gram matrix and the entropy of the imaging image as the objective function. By comparison, it can be seen that when imaging is performed using the emission waveform and the position of the emission element obtained by the present invention, the imaging effect is the best and clearest, the energy dissipation of the target grid is the lowest, and the imaging quality is the best.

[0087] The above is only a preferred embodiment of the present invention. It should be pointed out that for ordinary technicians in this technical field, several improvements and modifications can be made without departing from the principle of the present invention. These improvements and modifications should also be regarded as the scope of protection of the present invention.

Claims

1. A method for designing radiation fields of multi-view MIMO radar correlation imaging based on image entropy, characterized in that: The following steps are involved: (1) Set up the MIMO radar correlation imaging scenario to generate the transmit signal waveform and transmit array element position. (2) According to the waveform of the transmitted signal, the position of the transmitting and receiving array elements, and the position of the imaging unit, N reference matrices A are obtained. n , concatenate the N reference matrices in the time dimension to obtain the expanded-dimensional reference matrix A = [A1; A2; ···; A n ]; statistical modeling is performed on each resolution unit of the target. After the transmitted signal is scattered by each resolution unit, it is sampled at the receiving end for J time to obtain N echo vectors y = [y1; y2; ···; y n ], and the expanded-dimensional echo vector y is obtained after splicing; (3) Based on the sparse reconstruction multi-view MIMO radar correlation imaging algorithm, the expanded-dimensional reference matrix and the expanded-dimensional echo are jointly processed to obtain the scattering intensity value of each resolution unit of the imaging scene; (4) Set the maximum number of iterations Z max , taking the radiation field Gram matrix condition number and the image entropy obtained from the scattering intensity value of each resolution unit of the imaging scene as the objective function, the waveform of the transmitting signal and the formation of the transmitting array are iteratively optimized by constraining the radiation field Gram matrix condition number and image entropy; (5) If the number of iterations has not been reached, return to step (2); if the number of iterations has been reached, output the optimal signal waveform and the formation of the transmitting array.

2. The multi-view MIMO radar correlation imaging radiation field design method based on image entropy according to claim 1 is characterized in that: The scene setting implementation process of step (1) is as follows: The number of transmitting array elements of MIMO radar is M, the number of receiving array elements is N, the transmitting array is a random linear array, the receiving array is a uniform linear array, and the transmitting signal uses a random frequency hopping signal; the imaging plane is divided into L = x num *y num Imaging units, x num represents the number of lateral imaging units, y num represents the number of longitudinal imaging units, and the number of time sampling is J.

3. The multi-view MIMO radar correlation imaging radiation field design method based on image entropy according to claim 1 is characterized in that: The transmission signal waveform described in step (1) is: Among them, the waveform of the random frequency-hopping signal transmitted by the m-th transmitting array element is S m (), which consists of Q sub-pulses. q represents the q-th sub-pulse, and f c represents the carrier frequency, Δt and Δf represent the pulse duration and the minimum frequency interval respectively, G is a positive integer, and c m,q is the frequency-hopping code corresponding to each sub-pulse, and c m,q ∈{0, 1, …, G - 1}; u(t) represents a gate function that takes the value 1 when 0 < t < Δt and 0 otherwise.

4. The multi-view MIMO radar correlation imaging radiation field design method based on image entropy according to claim 1 is characterized in that: The transmitting array element position described in step (1) is: The position of the first element in the transmitting array is p1 = 0, and the position of the last element is p M =F, except for the first and last array elements, the positions of other array elements p m ∈{1,2,···F-1},m=2,3,···M-1; F is a positive integer; The position P of the transmitting array element to be optimized is expressed by the following formula: P={p m |p m ∈{1,2,…F-1},if i≠j then p i ≠p j ,m,i,j=2,3,…M-1} (3)。 5. The method for designing radiation field of multi-view MIMO radar correlation imaging based on image entropy according to claim 1, characterized in that: The process of implementing the statistical modeling of each resolution unit of the target in step (2) is as follows: The scattering intensity σ of each resolution unit is statistically modeled using a complex Gaussian model, σ = A0e jφ , σ obeys complex Gaussian distribution, the amplitude A0 of σ obeys Rayleigh distribution, and the phase φ of σ obeys uniform distribution: Among them, f(A0) is the probability density of amplitude, f(φ) is the probability density of phase, A n (r l ,t) is the reference signal, y n (t) is the echo signal, S m (t) represents the transmission signal of the mth transmitting array element, c is the propagation speed of electromagnetic waves, is the path delay from the mth transmitting array element through the lth imaging unit to the nth receiving array element; is the scattering intensity of the lth imaging unit at the viewing angle from the mth transmitting element to the nth receiving element; w n (t) is the noise of the nth receiving channel.

6. The method for designing radiation field of multi-view MIMO radar correlation imaging based on image entropy according to claim 1, characterized in that: The step (3) is implemented by the following formula: Among them, ‖·‖1 represents the l1 norm of the vector, ‖·||2 represents the l2 norm of the vector, and μ is the regularization parameter, which is used to control the balance between the l1 regularization term and the data fitting term. By minimizing the objective function, the scattering vector of the imaging target that meets the sparse condition is obtained.

7. The method for designing radiation field of multi-view MIMO radar correlation imaging based on image entropy according to claim 1, characterized in that: The step (4) is implemented by the following formula: Among them, A is the expanded radiation field reference matrix, A H represents the conjugate transpose of A, cond(·) represents the condition number, h(v) represents the image entropy, and v l represents the estimated value of the scattering intensity of each imaging unit, ||·||1 represents the l1 norm of the vector, ξ represents the optimization weight, C is the frequency hopping coding matrix, and P is the non-repeated transmitting array element position coding set.

8. The method for designing radiation field of multi-view MIMO radar correlation imaging based on image entropy according to claim 1, characterized in that: The iterative optimization solution described in step (4) is implemented by the gray wolf optimization algorithm, and the process is as follows: The GWO algorithm models the encirclement behavior using the following equations: D=|C'·X p (z)-X(z)| (10) X(z+1)=X p (z)-E·D (11) E=2a·ε1-a (12) C'=2·ε2 (13) Among them, z represents the current iteration number, X p (z) is the position vector of the prey, X(z) represents the position vector of the gray wolf, a is the convergence factor, ε1 and ε2 are random vectors between [0,1]; The GWO algorithm simulates hunting behavior using the following equation: D α =|C′1·X α -X| D β =|C′2·X β -X| D δ =|C′3·X δ -X| (14) X1=X α -E1·D α X2=X β -E2·D β X3=X δ -E3·D δ (15) Each set of objective functions obtained by different transmitting waveforms and transmitting array element positions is called an individual, and α, β, and δ wolves are the three individuals with the best, second best, and third best performance, D α , D β , D δ That is, the gap between the remaining individuals and the top three individuals, X α , X β , X δ That is, the transmission waveforms and array element positions of the three individuals with the top three performances, C'1, C'2, and C'3 are vectors randomly generated in the process, X is the transmission signal waveform and transmission array formation of the individual, and X(z+1) is the transmission signal waveform and transmission array formation corresponding to the individual after update.

Images