A Sparse Extended Target Imaging Method Based on a Metasurface Microwave Correlation Imaging System

Abstract

The present invention relates to a sparse extended target imaging method based on a metasurface microwave correlation imaging system, comprising the following steps: S1, constructing a metasurface microwave correlation imaging system; S2, signal modulation and transmission; S3, spatial setting of the imaging plane; S4, constructing a signal model; S5, constructing an imaging equation; S6, structuring and reorganizing a reference matrix block; S7, solving a computational imaging reconstruction algorithm. The present invention has the advantage of good imaging effect.

Description

Technical Field

[0001] The present invention belongs to the technical field of target imaging, and particularly relates to a sparse extended target imaging method based on a metasurface microwave correlation imaging system. Background Art

[0002] Microwave correlation imaging is one of the new radar imaging technologies, which can break through the limitation of antenna aperture on imaging resolution and has advantages such as forward-looking, staring, and snapshot imaging. In recent years, the rapidly developing metasurface technology has provided a new way for microwave correlation imaging. In reality, the sparse extension characteristics of targets widely exist in fields such as speech signal processing, medical image processing, and radar image processing. The design of imaging reconstruction algorithms is the core link of metasurface microwave correlation imaging.

[0003] At present, there are many theories and algorithms for sparse point target reconstruction based on microwave correlation imaging systems. However, the research theories and algorithms for sparse extended targets are relatively few. Domestic and foreign research mainly uses methods such as orthogonal matching pursuit (OMP) and sparse Bayesian learning (SBL) to solve microwave correlation imaging problems. However, these two existing methods do not consider the correlation within sparse extended targets and intend to use the sparse point target imaging framework to solve the sparse extended target imaging problem, resulting in defects such as low imaging accuracy and long imaging time for sparse extended targets. Therefore, it is of great practical value to design a new reconstruction method for sparse extended target imaging. Summary of the Invention

[0004] The purpose of the present invention is to overcome the deficiencies of the prior art and provide a sparse extended target imaging method based on a metasurface microwave correlation imaging system with good imaging effects.

[0005] The technical solution of the present invention is as follows:

[0006] A sparse extended target imaging method based on a metasurface microwave correlation imaging system includes the following steps:

[0007] Step S1: Construct a metasurface microwave correlation imaging system consisting of a transmitting antenna, a reflective two-bit phase modulation metasurface, and a receiving antenna;

[0008] Step S2: Modulate the transmission signal of the transmitting antenna and then transmit it through space to act on the target. After scattering, the transmission signal is received by the receiving antenna;

[0009] Step S3: Set an imaging plane directly in front of the reflective two-bit phase modulation metasurface. The imaging plane is evenly divided into square grids of the same size, and the scattering points are located at the grid centers of the imaging plane. The normal line of the reflective two-bit phase modulation metasurface passes through the plane center of the imaging plane;

[0010] Step S4: Construct a signal model to obtain the reflected signal and the reference signal ;

[0011] Step S5: Construct an imaging equation to obtain the received signal vector Sr and the reference matrix ;

[0012] Step S6: Reconstruct the reference matrix block by block, and finally obtain a new reference matrix containing the imaging block structure information , create a structure information index Sindex Record the index of each atom in the new reference matrix in the reference matrix ;

[0013] Step S7: Calculate the solution of the imaging reconstruction algorithm, and the specific steps are as follows:

[0014] Input: Received signal Sr , reference matrix , the block sparsity of the signal K , structure information index Sindex ;

[0015] Output: Reconstructed scene scattering intensity ;

[0016] Step S71: Initialize the BOMP algorithm: Set the block iteration number k =0, the residual r = Sr , the selected atom set Φ k is an empty set, and the recovery index set Λ k is an empty set;

[0017] Step S72: Iterative step, the k th iteration;

[0018] (1) Find the relevant block r that best matches the residual S in the reference matrix D j among all relevant blocks D j , and the specific formula is as follows; ;

[0019] (2) Record the index set D j of the positive atoms in the relevant block in the new reference matrix J k , increase the recovery index set , and increase the support set ;

[0020] (3) Least - squares reconstruction of the target signal , where represents the pseudo - inverse of matrix ;

[0021] (4) Update the residual ;

[0022] (5) Determine whether the number of loops reaches the set number of loops or the residual norm is less than the set threshold. If satisfied, stop; if not, continue the iteration;

[0023] Step S73: According to the recovery index set Λ k , the structure information index Sindex and , match the scattering coefficients of the reconstruction space with the indices to obtain the target microwave correlation imaging result.

[0024] Further, the specific process of constructing the signal model in step S4 is as follows:

[0025] Assume that the number of units of the reflective two - bit phase - modulation metasurface is N , the number of grids in the imaging plane is L , the starting time of the signal at the transmitting end is t , the transmitted signal is S(t) , the distance from the transmitting antenna to the n - th unit of the reflective two - bit phase - modulation metasurface is , the distance from the n - th unit to the - th imaging grid is , the distance from the - th imaging unit to the receiving antenna is . Assume that the signal irradiated by the transmitting antenna on the metamaterial surface is a plane wave, that is, the signals received by each metamaterial unit have equal amplitudes. The incident signal n propagating to the - th unit of the metasurface is expressed as: ;

[0026] where c represents the speed of light. The outgoing signal n of the - th unit of the metamaterial is expressed as:

[0027] ;

[0028] where is the random phase shift amount of the two - bit metamaterial unit, takes values 0 or π , j is the imaginary unit, and all NThe outgoing signal of a unit reaches the imaging grid space point The superimposed incoming signal Is expressed as:

[0029] ;

[0030] After acting on the target, the outgoing signal from the grid space point Is expressed as: Is expressed as: ;

[0031] Where Represents the target scattering coefficient at the spatial grid point The signal received by the receiving antenna Sr(t) Is expressed as the superposition of the outgoing signals of all grid points as: ;

[0032] Select a random frequency hopping signal as the transmitted signal, then the transmitted signal S(t) The specific form is: ;

[0033] Where Is the center frequency, Is the minimum frequency hopping interval, Is the frequency hopping control code, that is, the total delay of the signal received by the receiving antenna : ;

[0034] Then the signal received by the receiving antenna Sr(t) Is expressed as: ;

[0035] The Reference signal of the imaging grid point Is expressed as: .

[0036] Furthermore, the specific process of constructing the imaging equation in step S5 is as follows:

[0037] Discretize the transmitted signal selected from the random frequency hopping signal S(t) And design the timing of the random frequency hopping signal and the random coding sequence configuration, perform single-point sampling within each coding time, and one coding time represents one measurement. Assume the coding time is M , construct the microwave correlation imaging equation set as:

[0038] ;

[0039] Where Represents theM The received signal under a coding, representing the reference signal of the M -th grid under the L -th coding, being the target scattering intensity of the L -th grid, being the noise of the L -th grid, the matrix representation of the microwave correlation imaging equations is as follows:

[0040] Sr = S · + n

[0041] Wherein Sr is the received signal vector, S is the reference matrix, is the target scattering coefficient vector, n is the additive noise vector.

[0042] Furthermore, the specific steps of the reference matrix block-structured recombination in step S6 are as follows:

[0043] SS301. Divide the scene D into several square imaging blocks D = {D1,... D M};

[0044] S302. Place the atoms of the corresponding reference matrix S together column by column according to the neighbor characteristics of the imaging grid structure within the two-dimensional square block;

[0045] S303. Obtain a new reference matrix containing the imaging block structure information, and create a structure information index Sindex to record the index of each atom in the new reference matrix in the reference matrix S.

[0046] Compared with the prior art, the beneficial effects of the present invention are:

[0047] 1. The present invention constructs the constraint of block reconstruction through step S7. The selection of atoms in the orthogonal matching pursuit algorithm depends on the level of the overall block correlation strength, ensuring that the neighborhood information of point targets can also be reconstructed, improving the probability of reconstructing the target center, reducing the possibility of being interfered by strong scattering points, and improving the robustness of the algorithm.

[0048] 2. The present invention adopts the positive correlation atom addition strategy through step S7, so that only atoms showing positive correlation can be added to the reconstruction support set. This strategy conforms to the principle of microwave correlation imaging, improves the accuracy while reducing the computational amount of the algorithm and shortening the imaging time. BRIEF DESCRIPTION OF THE DRAWINGS

[0049] Figure 1 is a schematic diagram of the metasurface microwave correlation imaging system of the present invention;

[0050] Figure 2 It is the flow chart of the imaging equation reconstruction algorithm of the present invention;

[0051] Figure 3 It is the target scene diagram of the embodiment of the present invention;

[0052] Figure 4 It is the target scene imaging reconstruction result diagram provided by the embodiment of the present invention; Detailed implementation manners

[0053] Next, the technical solutions in the embodiments of the present invention will be clearly and completely described in conjunction with the accompanying drawings in the embodiments of the present invention. Obviously, the described embodiments are only a part of the embodiments of the present invention, rather than all the embodiments. All other embodiments obtained by those of ordinary skill in the art based on the embodiments of the present invention without creative efforts shall fall within the protection scope of the present invention.

[0054] As Figures 1 - 4 shown, a sparse extended target imaging method based on a metasurface microwave correlation imaging system specifically includes the following steps:

[0055] It includes the following steps:

[0056] Step S1: Construct a metasurface microwave correlation imaging system composed of a transmitting antenna, a reflective two-bit phase modulation metasurface, and a receiving antenna. The metamaterial units on the reflective two-bit phase modulation metasurface are evenly distributed in a plane, and the distance between adjacent units is half a wavelength;

[0057] Step S2: The transmitting antenna emits a random frequency hopping signal. The signal is incident on the reflective two-bit phase modulation metasurface with random phase modulation, generating a space-time uncorrelated random radiation field. After the signal is modulated, it is transmitted through space and acts on the target. The signal at the target is scattered and then received by the receiving antenna;

[0058] Step S3: Set an imaging plane directly in front of the reflective two-bit phase modulation metasurface. The imaging plane is evenly divided into square grids of the same size, and the scattering points are located at the grid centers of the imaging plane. The normal line of the reflective two-bit phase modulation metasurface passes through the plane center of the imaging plane;

[0059] Step S4: Construct a signal model to obtain the reflected signal and the reference signal ;

[0060] The specific process of constructing the signal model in Step S4 is as follows:

[0061] Assume that the number of units of the reflective two-bit phase modulation metasurface is N , and the number of grids of the imaging plane isL , the starting time of the transmitting - end signal is t , the transmitted signal is S(t) , the distance from the transmitting antenna to the n th unit of the reflective two - bit phase - modulation metasurface , the n th unit to the th imaging grid distance , the th imaging unit to the receiving - antenna distance , assuming that the signal irradiated by the transmitting antenna on the metamaterial surface is a plane wave, that is, the signals received by each metamaterial unit have equal amplitudes. The incident signal of the transmitted signal propagating to the n th unit of the metasurface is expressed as: ;

[0062] where c represents the speed of light. The outgoing signal of the n th unit of the metamaterial is expressed as:

[0063] ;

[0064] where is the random phase - shift amount of the two - bit metamaterial unit, takes the value of 0 or π , j is the imaginary unit. The superposition of the outgoing signals of all N units of the metamaterial to the imaging - grid spatial point is expressed as: is expressed as:

[0065] ;

[0066] After acting on the target, the outgoing signal from the grid spatial point is expressed as: is expressed as: ;

[0067] where represents the target scattering coefficient at the spatial grid point . The signal received by the receiving antenna Sr(t) is expressed as the superposition of the outgoing signals of all grid points: ;

[0068] If a random frequency - hopping signal is selected as the transmitted signal, then the transmitted signal S(t) has the specific form: ;

[0069] where is the center frequency, is the minimum frequency hopping interval, is the frequency hopping control code, that is, the total delay of the signal received by the receiving antenna : ;

[0070] Then the signal received by the receiving antenna Sr(t) is expressed as: ;

[0071] The reference signal of the th imaging grid point .

[0072] Step S5: Construct an imaging equation to obtain the received signal vector Sr and the reference matrix ;

[0073] The specific process of constructing the imaging equation in step S5 is as follows:

[0074] Discretize the transmitted signal S(t) selected from the random frequency hopping signal, and design the timing of the random frequency hopping signal and the random coding sequence configuration. Perform single-point sampling within each coding time. One coding count represents one measurement. Assume the coding count is M , and construct the microwave correlation imaging equation set as:

[0075] ;

[0076] where represents the received signal under the M th coding, represents the reference signal of the M th grid under the L th coding, is the target scattering intensity of the L th grid, is the noise of the L th grid. The matrix representation of the microwave correlation imaging equation set is as follows:

[0077] Sr = S · + n

[0078] where Sr is the received signal vector, S is the reference matrix, is the target scattering coefficient vector, n is the additive noise vector.

[0079] Step S6: Structured recombination of the reference matrix blocks, and finally a new reference matrix containing imaging block structure information is obtained , create a structure information index Sindex Record the new reference matrix The index of each atom in the reference matrix ;

[0080] The specific steps of the structured recombination of the reference matrix blocks in step S6 are as follows:

[0081] SS301: Divide the scene D into several square imaging blocks D = {D1,... D M};

[0082] S302: Place the atoms of the corresponding reference matrix S together by column according to the near-neighbor characteristics of the imaging grid structure within the two-dimensional square block;

[0083] S303: Obtain a new reference matrix containing imaging block structure information , create a structure information index Sindex Record the index of each atom in the new reference matrix in the reference matrix S.

[0084] Step S7: Solve by calculating the imaging reconstruction algorithm. The present invention provides an improved block orthogonal matching pursuit (BOMP) algorithm to solve the imaging equation. As shown in the flowchart attached Figure 2 , the specific steps of the improved block orthogonal matching pursuit (BOMP) algorithm are as follows:

[0085] Input: Received signal Sr , reference matrix , block sparsity of the signal K , structure information index Sindex ;

[0086] Output: Reconstructed scene scattering intensity ;

[0087] Step S71: Initialization of the BOMP algorithm: Set the block iteration number k = 0, residual r = Sr , selected atom set Φ k is an empty set, and the recovery index set Λ k is an empty set;

[0088] Step S72: Iterative step, the k -th iteration;

[0089] (1) Find the residual r and the most matching relevant block S in the reference matrix D j among all relevant blocksD j The index in it is as follows; ;

[0090] (2)Record relevant blocks D j The positive atoms in the new reference matrix Index set in J k , increase the recovery index set , increase the support set ;

[0091] (3)Least squares reconstruction of the target signal , where Represents the pseudo-inverse of the matrix ;

[0092] (4)Update the residual ;

[0093] (5)Judge whether the number of loops reaches the set number of loops or the residual norm is less than the set threshold. If satisfied, stop. If not satisfied, continue the iteration;

[0094] Step S73: According to the recovery index set Λ k , structural information index Sindex and , match the scattering coefficients of the reconstructed space with the index to obtain the target microwave correlation imaging result.

[0095] The experimental imaging target scene shows the word "cyber" as attached Figure 3 , the imaging reconstruction result of the improved block orthogonal matching pursuit (BOMP) algorithm is as attached Figure 4 , the analysis result shows that a sparse extended target imaging method based on a metasurface microwave correlation imaging system proposed in this paper presents a good imaging effect.

[0096] Although the present invention has been described in detail with reference to the foregoing embodiments, for those skilled in the art, they can still modify the technical solutions described in the foregoing embodiments, or perform equivalent replacements for some of the technical features. Any modifications, equivalent replacements, improvements, etc. made within the spirit and principle of the present invention shall be included within the protection scope of the present invention.

Claims

1. A sparse extended target imaging method based on a metasurface microwave correlation imaging system, characterized in that, It includes the following steps: Step S1: Construct a metasurface microwave correlation imaging system composed of a transmitting antenna, a reflective two-bit phase modulation metasurface, and a receiving antenna; Step S2: Modulate the transmission signal of the transmitting antenna and then transmit it through space to act on the target. After scattering, the transmitted signal is received by the receiving antenna; Step S3: Set an imaging plane directly in front of the reflective two-bit phase modulation metasurface. The imaging plane is evenly divided into square grids of the same size, and the scattering points are located at the grid centers of the imaging plane. The normal line of the reflective two-bit phase modulation metasurface passes through the plane center of the imaging plane; Step S4: Construct a signal model to obtain the reflected signal and the reference signal ; Step S5: Construct an imaging equation to obtain a received signal vector Sr and a reference matrix ; Step S6: Structured recombination of the reference matrix blocks, finally obtaining a new reference matrix containing the imaging block structure information , creating a structure information index Sindex Recording the new reference matrix of each atom in the reference matrix index in; Step S7: Solve by calculating the imaging reconstruction algorithm. The specific steps are as follows: Input: Received signal Sr , reference matrix , block sparsity of the signal K , structure information index Sindex ; Output: Reconstructed scene scattering intensity ; Step S71: Initialization of the BOMP algorithm: Set the number of block iterations k = 0, the residual r = Sr , select the atom set Φ k is an empty set, restore the index set Λ k is an empty set; Step S72: Iterative step, the k th iteration; (1) Finding the residual r with the reference matrix S the most matching relevant block D j among all relevant blocks D j the index in, and the specific formula is as follows; ; (2)Record relevant blocks D j The index set of positive atoms in the new reference matrix in J k , increase the recovery index set , increase the support set ; (3) Least squares reconstruction of the target signal where represents the pseudo-inverse of the matrix ; (4) Update the residual ; (5) Determine whether the number of loop iterations has reached the set number of loop iterations or the residual norm is less than the set threshold. If satisfied, stop; if not satisfied, continue the iteration; Step S73: According to the recovery index set Λ k , the structure information index Sindex and , match the scattering coefficient of the reconstructed space with the index to obtain the target microwave correlation imaging result.

2. The sparse extended target imaging method based on a metasurface microwave correlation imaging system according to claim 1, wherein: The specific process of constructing the signal model in step S4 is as follows: Assume that the number of cells of the reflective two-bit phase modulation metasurface is N , the number of grids in the imaging plane is L , the starting time of the signal at the transmitting end is t , the transmitted signal is S(t) , the distance from the transmitting antenna to the n -th cell of the reflective two-bit phase modulation metasurface is , the distance from the n -th cell to the -th imaging grid is , the distance from the -th imaging unit to the receiving antenna is , assume that the signal irradiated by the transmitting antenna on the metamaterial surface is a plane wave, that is, the signals received by each metamaterial cell have equal amplitude. The incident signal of the transmitted signal propagating to the n -th cell of the metasurface is expressed as: ; where c represents the speed of light, and the output signal of the n th unit of the metamaterial is expressed as: ; wherein is the random phase shift of the two-bit metamaterial unit, taking values of 0 or π , j is the imaginary unit, and the superposition of the outgoing signals of all N units of the metamaterial to the imaging grid space point of the incident signal is expressed as: ； After acting on the target, the outgoing signal from the grid space point is expressed as: ； wherein represents the target scattering coefficient at the spatial grid point and the signal received by the receiving antenna Sr(t) is expressed as the superposition of the outgoing signals at all grid points as: ; Select a random frequency hopping signal as the transmitted signal, then the transmitted signal S(t) The specific form is: ； wherein is the center frequency, is the minimum frequency hopping interval, is the frequency hopping control code, i.e., the total delay of the signal received by the receiving antenna , and the is the amplitude of the signal: ; Then the receiving antenna receives the signal Sr(t) It is expressed as: ； The reference signal of the imaging grid points is expressed as: 。 3. A sparse extended target imaging method for a metasurface microwave correlation imaging system according to claim 2, characterized in that: The specific process of constructing the imaging equation in step S5 is as follows: The transmitted signal selected from the random frequency-hopping signal S(t) is discretized, and the timing sequence of the random frequency-hopping signal and the random coding sequence is designed. Single-point sampling is performed within each coding time, and one coding count represents one measurement. Assuming the coding count is M , the microwave correlation imaging equation set is constructed as follows: ； Among them represents the received signal under the M th code, represents the reference signal of the M th grid under the L th code, is the target scattering intensity of the L th grid, is the noise of the L th grid. The matrix representation of the microwave correlation imaging equation set is as follows: Sr = S· + n; wherein Sr is the received signal vector, S is the reference matrix, is the target scattering coefficient vector, n is the additive noise vector.

4. A sparse extended target imaging method for a metasurface microwave correlation imaging system according to claim 1, characterized in that: The specific steps of the reference matrix block structured recombination in step S6 are as follows: S301. Divide the scene D into a number of square imaging blocks D = {D1,...D M}; S302: Place the atoms of the corresponding reference matrix S together column by column according to the near-neighbor characteristics of the imaging grid structure within the two-dimensional square block; S303. Obtain a new reference matrix containing the imaging block structure information , create a structure information index Sindex Record the new reference matrix The index of each atom in the reference matrix S

Images