Array signal sparse reconstruction method and device based on virtual array fragment expansion

Abstract

The invention relates to the technical field of radar signal processing, in particular to an array signal sparse reconstruction method and device based on virtual array fragment expansion. The method mainly comprises the following steps: generating an extended co-prime physical array based on configuration parameters according to array element requirements; constructing a far-field signal of the extended co-prime physical array; modeling an extended co-prime array receiving signal, constructing an extended co-prime array virtual array, and performing continuity judgment; for the continuous virtual array, target orientation estimation of the uniform linear array is carried out based on the continuous virtual array; for a discontinuous virtual array, two sections with the longest length in virtual array fragments are selected, signal modeling is performed according to the longer section of array, and the other section of array signal is selected as a verification array, so that discontinuous virtual array signal reconstruction based on virtual array fragment expansion is realized. The discontinuous virtual array signal sparse reconstruction based on virtual array fragment expansion solves the problem that array element damage is difficult to repair in the underwater acoustic array laying process.

Description

Technical Field

[0001] The present invention relates to the field of radar signal processing technology, and in particular to a method and device for sparse reconstruction of array signals based on virtual array segment expansion. Background Art

[0002] Coprime arrays have received extensive attention in radar because they can implement high-degree-of-freedom virtual array element domain signal processing algorithms. In recent years, research and discussion on coprime arrays in underwater signal processing has gradually begun in the field of underwater array signal processing. In underwater coprime array signal processing, signal processing methods based on the physical array element domain focus on achieving high-robustness target detection, while signal processing methods based on the virtual array element domain focus on achieving high-resolution target detection.

[0003] When performing virtual domain signal processing of coprime arrays, array expansion can be used to achieve a longer virtual array deployment, thereby achieving a large degree of freedom, which is of great significance in the field of radar target detection.

[0004] In underwater array signal processing, the complex underwater environment determines that it is difficult to ensure the stability of the target signal for a long time, which also puts forward the demand for small snapshots in signal processing. In addition, when the array element is damaged during the test, repairing it will bring huge and complex workload, which is also a major problem in practical applications. Summary of the invention

[0005] In view of the need for small snapshot processing in complex underwater environments and the huge and complex repair workload when array elements are damaged during the deployment of underwater acoustic arrays, the present invention provides a method and device for sparse reconstruction of array signals based on virtual array fragment expansion. The present invention also solves the problem that coprime arrays generate discontinuous virtual arrays, thereby affecting the target azimuth estimation performance of virtual arrays.

[0006] The technical means adopted by the present invention are as follows:

[0007] A method for sparse reconstruction of array signals based on virtual array segment expansion comprises the following steps:

[0008] S1. Obtaining hardware parameters of the initial array, the hardware parameters including the total number of array elements and the array aperture; obtaining nonlinear array-coprime array configuration parameters according to the hardware parameters, the configuration parameters including array expansion cardinality (M, N) and array expansion factors β of two sub-arrays;

[0009] S2, obtaining array element requirements, and expanding the coprime array in the physical array element domain based on the array element requirements and configuration parameters, thereby generating an extended coprime physical array;

[0010] S3, constructing the far-field signal of the extended mutually prime physical array;

[0011] S4, modeling the extended coprime array receiving signal, obtaining the steering matrix corresponding to each virtual array element in the virtual array, and constructing the extended coprime array virtual array based on parameter configuration based on the steering matrix corresponding to each virtual array element;

[0012] S5, judging the continuity of the extended coprime array virtual array, if the obtained extended coprime array virtual array is a continuous virtual array, executing S6, otherwise executing S7;

[0013] S6, target azimuth estimation of uniform linear array based on continuous virtual array;

[0014] S7. Segment the virtual array, select the two longest segments in the virtual array fragments, perform signal modeling based on the longer segment of the selected array, select another segment of the array signal as a verification array, and realize non-continuous virtual array signal reconstruction based on the virtual array fragments, and finally obtain the virtual array information and high-resolution target azimuth estimation result output.

[0015] Furthermore, the far-field signal of the extended coprime physical array is constructed, including:

[0016] The far-field narrowband signal is constructed as follows:

[0017] X(t)=[x1(t),x2(t),...x P (t)] T =Az(t)+N(t)

[0018] where z(t)=[z1(t),z2(t),…z L (t)] T is the target signal vector received by the array, L is the number of far-field narrowband signals, is a statistically independent Gaussian noise component, P is the number of array elements, I is the unit matrix, is the noise power,

[0019] A=[a(θ0),a(θ1),...,a(θ L )] is the array flow matrix, 1≤l≤L,a(θ l ) is the array flow pattern vector of the target in the array, is the distance between two adjacent array elements, where d2 is the distance between the first array element and the second array element, d3 is the distance between the second array element and the third array element, and d P is the distance between the P-1th array element and the Pth array element, λ is the signal wavelength, θ l is the signal incident azimuth;

[0020] When the processed signal is a broadband signal, the far-field broadband signal is constructed according to the following method:

[0021] The broadband signal is equally divided into several narrowband signals, and then the signal is constructed based on the construction method of the narrowband signal. The pre-examination processing frequency band range of the far-field broadband is [f L ,f H ].

[0022] Furthermore, modeling the extended coprime array received signal includes:

[0023] Construct the covariance matrix of the received signal of the extended coprime array virtual array based on parameter selection:

[0024]

[0025] Where T is the signal length, X(t) is the array receiving signal, and X H (t) is the conjugate transpose of the signal,

[0026] Vectorizing the covariance, we get:

[0027]

[0028] Among them, A v is the steering matrix corresponding to each virtual array element in the virtual array, u=vec(I), p=[p1,p2,...,p L ] T is the power of the incident signal.

[0029] Furthermore, the target azimuth of the uniform linear array is estimated based on the continuous virtual array, including:

[0030] Using the traditional robust beamforming method, a virtual uniform linear array flow vector is constructed to obtain the virtual array beam output and target azimuth information.

[0031] Furthermore, the target azimuth of the uniform linear array is estimated based on the continuous virtual array, including:

[0032] Using the high-resolution subspace target direction estimation method, a virtual uniform linear array signal covariance is constructed, and target direction estimation is achieved through signal and noise subspace processing.

[0033] Furthermore, the target azimuth of the uniform linear array is estimated based on the continuous virtual array, including:

[0034] Using the principle of compressed sensing, virtual uniform linear array target sparse recovery and target direction estimation are constructed.

[0035] Further, signal modeling is performed according to a longer segment of the selected array, and another segment of the array signal is selected as a verification array, including:

[0036] The signal vector is constructed as follows:

[0037]

[0038] Where V is the measurement vector in the virtual array element domain, V cv is the validation vector:

[0039]

[0040]

[0041] Where p and q represent the element positions of the virtual array, represents the element position in the entire continuous virtual array, D i represents the elements in the array fragment, i is the number of array fragments, let is the iterative result of the received signal covariance quantization, k is the number of iterations, and the sparse estimation result of the kth iteration of sparse reconstruction, then the verification residual can be obtained based on the virtual array fragment of this iteration. When the number of iterations increases, ε k becomes smaller; when the restored signal is overfitted, ε k becomes larger, when ε k The iteration stops when the minimum value is reached:

[0042]

[0043] Where E represents the maximum number of iterations.

[0044] The present invention also discloses an array signal sparse reconstruction device based on virtual array segment expansion, which is used to implement the above method, including:

[0045] A parameter acquisition unit, used to acquire hardware parameters of the initial array, the hardware parameters including the total number of array elements and the array aperture; acquire configuration parameters of a special array-coprime array according to the hardware parameters, the configuration parameters including an array expansion cardinality (M, N) and an array expansion factor β of two sub-arrays;

[0046] An extended coprime physical array construction unit is used to obtain array element requirements, and expand the coprime array in the physical array element domain based on the array element requirements and configuration parameters, thereby constructing an extended coprime physical array;

[0047] A signal construction unit, used for constructing a far-field signal of an extended mutually prime physical array;

[0048] An extended coprime array virtual array construction unit is used to model the extended coprime array receiving signal, obtain the steering matrix corresponding to each virtual array element in the virtual array, and construct an extended coprime array virtual array based on parameter configuration based on the steering matrix corresponding to each virtual array element;

[0049] A judging unit, used for judging the continuity of the extended coprime array virtual array;

[0050] A first target estimation unit, used for estimating the target position of a uniform linear array based on a continuous virtual array;

[0051] The second target estimation unit is used to segment the virtual array, select the two longest segments in the virtual array fragments, perform signal modeling based on the longer segment of the selected array, select another segment of the array signal as a verification array, and realize non-continuous virtual array signal reconstruction based on the virtual array fragments, and finally obtain the virtual array information and high-resolution target azimuth estimation result output.

[0052] The present invention also discloses a storage medium, which includes a stored program, wherein when the program is run, the method described in any one of the above claims is executed.

[0053] The present invention also discloses an electronic device, comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the processor executes the method described in any one of the above claims through the operation of the computer program.

[0054] Compared with the prior art, the present invention has the following advantages:

[0055] The present invention provides a method for realizing sparse reconstruction of array signals based on virtual array fragment expansion. The method combines the characteristics of coprime arrays and aims at the problem of virtual array breakage caused by expansion and array element failure in underwater signal processing applications of coprime arrays. A discontinuous virtual array signal sparse reconstruction method using virtual array fragment expansion is designed. On the one hand, the demand for small snapshot data processing in underwater array signal processing is met. On the other hand, when an array element fails, array information can still be fully utilized to realize high-resolution target azimuth estimation. BRIEF DESCRIPTION OF THE DRAWINGS

[0056] In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the drawings required for use in the embodiments or the description of the prior art will be briefly introduced below. Obviously, the drawings described below are some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying creative labor.

[0057] Figure 1It is a flow chart of the method of the present invention.

[0058] Figure 2 It is a schematic diagram of the geometric structure of the physical array element domain and virtual array element domain of the coprime array.

[0059] Figure 3 This is the result of sparse reconstruction of the discontinuous virtual domain signal obtained by expansion.

[0060] Figure 4 This is a schematic diagram of a discontinuous virtual array caused by array element damage.

[0061] Figure 5 This is the result of sparse reconstruction of non-continuous virtual domain signals.

[0062] Figure 6 It is a graph showing how the algorithm MSE changes with the signal-to-noise ratio and the number of snapshots, where (a) is a graph showing how the MSE changes with the signal-to-noise ratio, and (b) is a graph showing how the MSE changes with the number of snapshots.

[0063] Figure 7 This is a schematic diagram showing errors in array element positions.

[0064] Figure 8 The following are the algorithm result diagrams when there is an array element position error, where (a) is the target orientation of 20° and 24°, and (b) is the target orientation of 20° and 23°. DETAILED DESCRIPTION

[0065] In order to enable those skilled in the art to better understand the scheme of the present invention, the technical scheme in the embodiments of the present invention will be clearly and completely described below in conjunction with the drawings in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work should fall within the scope of protection of the present invention.

[0066] It should be noted that the terms "first", "second", etc. in the specification and claims of the present invention and the above-mentioned drawings are used to distinguish similar objects, and are not necessarily used to describe a specific order or sequence. It should be understood that the data used in this way can be interchanged where appropriate, so that the embodiments of the present invention described herein can be implemented in an order other than those illustrated or described herein. In addition, the terms "including" and "having" and any variations thereof are intended to cover non-exclusive inclusions, for example, a process, method, system, product or device that includes a series of steps or units is not necessarily limited to those steps or units that are clearly listed, but may include other steps or units that are not clearly listed or inherent to these processes, methods, products or devices.

[0067] like Figure 1 As shown, the present invention provides an array signal sparse reconstruction method based on virtual array segment expansion, comprising the following steps:

[0068] S1. Obtain hardware parameters of the initial array, the hardware parameters including the total number of array elements and the array aperture; obtain special array configuration parameters of coprime array according to the hardware parameters, the configuration parameters including array expansion cardinality (M, N) and array expansion factor β of two sub-arrays.

[0069] For an initial array, set the parameters required for the special array - coprime array configuration according to its hardware parameters for the array (total number of array elements, array aperture, etc.), and set the selection constraints for the array expansion cardinality (M, N) and the array expansion factor β:

[0070] ①M and N are a pair of coprime numbers, ensuring that their greatest common divisor is 1;

[0071] ② The traditional coprime array is composed of two uniform subarrays. Therefore, there are many ways to expand the two subarrays. The present invention sets the subarray expansion mode to ensure that the two subarrays are expanded with the same array expansion factor β, that is:

[0072] β M =β N

[0073] In the expanded coprime array physical element domain, the total theoretical number of elements P is:

[0074] P=β(M+N)-1

[0075] β=β M =β N Due to the existence of the expansion factor constraint, the total number of array elements in the actual expanded array will be P'≤P. This is because the particularity of the expansion factor will make it possible for the position of repeated array elements to exist between the expanded mutual proton arrays. The above formula guarantees the upper limit of the total number of array elements in the expanded array. After obtaining the theoretical total number of array elements using the above formula, the total number of array elements needs to be further confirmed based on the actual array layout.

[0076] S2. Obtain array element requirements, and expand the coprime array in the physical array element domain based on configuration parameters according to the array element requirements, thereby generating an expanded coprime physical array.

[0077] Generally, when the total number of array elements is fixed to O, the coprime array constructed based on (M, N) satisfies O=M+N-1. The coprime array expansion scheme is usually adopted to achieve a large array aperture with a limited number of array elements. When the array aperture is fixed, the use of an extended coprime array can fully utilize the array aperture space, save the array hardware burden and improve the spatial resolution. The present invention is based on the virtual array domain expansion generated by the extended coprime array. Figure 2 As shown in the figure, a comparison diagram of the array structure in signal processing in the physical array element domain and the virtual array element domain is shown.

[0078] S3. Construct the far-field signal of the extended coprime physical array.

[0079] The coprime array is expanded using the selected expansion factor to achieve the construction of L far-field narrowband signals at time t of the extended coprime physical array based on special sequence pairs:

[0080] X(t) = [x1(t), x2(t), L x P (t)] T =Az(t)+ N(t) (2)

[0081] The variables are defined as follows:

[0082] z(t)=[z1(t),z2(t),L z L (t)] T receiving a target signal vector for the array;

[0083] is a statistically independent Gaussian noise component; I is the unit matrix, is the noise power;

[0084] A=[a(θ0),a(θ1),...,a(θ L )] is the array flow matrix, a(θ l ) is the array flow vector of the target in the array, d2, d3, ...d P is the distance between two adjacent array elements, where d2 is the distance between the first array element and the second array element, d3 is the distance between the second array element and the third array element, and d P is the distance between the P-1th array element and the Pth array element, λ is the signal wavelength, θ l is the signal incident direction.

[0085] For broadband signals, such as bandwidth f B , then the broadband is divided into multiple narrowbands [f V ,f 2V ,...,f B ], and then model the narrowband signal in the above way.

[0086] S4, modeling the extended coprime array receiving signal, obtaining the steering matrix corresponding to each virtual array element in the virtual array, and constructing the extended coprime array virtual array based on parameter configuration based on the steering matrix corresponding to each virtual array element;

[0087] The extended coprime array received signal is modeled and the extended coprime array virtual array received signal covariance matrix based on parameter selection is constructed:

[0088]

[0089] Vectorizing the covariance, we get:

[0090]

[0091] in, u=vec(I), p=[p1,p2,...,p L ] T . A v It is the steering matrix corresponding to each virtual array element in the virtual array.

[0092] S5. Perform continuity judgment on the extended coprime matrix virtual array. If the obtained extended coprime matrix virtual array is a continuous virtual array, execute S6; otherwise, execute S7.

[0093] Examine the obtained virtual arrays and establish the difference set of coprime arrays:

[0094]

[0095] S6. Target azimuth estimation of uniform linear array based on continuous virtual array.

[0096] If a continuous virtual array is obtained, the target position estimation of the virtual array as a uniform linear array is performed according to the requirements:

[0097] Optionally, a conventional robust beamforming method, such as CBF (Conventional Beamforming) and Capon, is used to construct a virtual uniform linear array flow vector to obtain a virtual array beam output and target azimuth information;

[0098] Optionally, a high-resolution subspace target azimuth estimation method, such as MUSIC (Multiple Signal Classification), is used to construct a virtual uniform linear array signal covariance, and target azimuth estimation is achieved through signal and noise subspace processing;

[0099] Optionally, the compressed sensing principle is used to construct virtual uniform linear array target sparse recovery and target direction estimation.

[0100] S7. Segment the virtual array, select the two longest segments in the virtual array fragments, perform signal modeling based on the longer segment of the selected array, select another segment of the array signal as a verification array, and realize non-continuous virtual array signal reconstruction based on the virtual array fragments, and finally obtain the virtual array information and high-resolution target azimuth estimation result output.

[0101] When the obtained virtual array is a non-continuous virtual array, the virtual array is segmented and the two longest segments of the virtual array fragments are selected. When a non-continuous virtual array appears, the following reasons may exist:

[0102] ① After expansion, the obtained virtual array has array "holes";

[0103] ② When the physical array elements are damaged or malfunction, “holes” will appear in the virtual array.

[0104] Based on the above two situations, "holes" will be generated in the array, interrupting the originally continuous virtual array, resulting in an incomplete array, making the array signal processing method based on uniform continuity unapplicable, and causing a waste of resources of virtual array elements in the interrupted array segment. In view of this situation, the present invention designs an algorithm for this situation. First, the "holes" in the virtual array divide a continuous virtual array into multiple parts. At this time, the position of the virtual array element can be expressed as:

[0105]

[0106] in, represents the element position in the entire continuous virtual array, D i Represents the elements in the array fragments obtained by "hole" segmentation. Here, i is the number of array fragments.

[0107] Signal modeling is performed based on a longer segment of the selected array, and another segment of the array signal is selected as a verification array to achieve non-continuous virtual array signal reconstruction based on virtual array fragment expansion, and finally obtain virtual array information and high-resolution target direction estimation result output. The present invention uses sparse signal reconstruction technology to perform sparse DOA (Direction Of Arrival) estimation. The signal corresponding to the discontinuous virtual array is divided into two parts, and a compressed sensing technology based on virtual array fragment verification is used.

[0108]

[0109]

[0110] Where V is the measurement vector in the virtual array element domain, V cvIt is the verification vector in MMP (Multipath Matching Pursuit, MMP):

[0111]

[0112] set up is the sparse estimation result of the kth iteration of sparse reconstruction, then the verification residual can be obtained based on the virtual array fragment of this iteration. When the number of iterations increases, ε k becomes smaller; when the restored signal is overfitted, ε k When ε k The iteration stops when the minimum value is reached:

[0113]

[0114] E represents the maximum number of iterations.

[0115] The present invention verifies the performance of the method through the following simulation experiments.

[0116] Simulation experiment 1

[0117] like Figure 3 As shown, the simulation analysis realizes the target azimuth estimation based on the sparse reconstruction method of non-continuous virtual array signals by expanding virtual array fragments. The simulation conditions are M=5, N=7, and the array expansion factor β=2. At this time, the generated virtual array will produce array "holes", and the continuity of the virtual array will be interrupted, resulting in the generation of array fragments. In the generated non-continuous virtual array, the array data of 93 array elements with continuous virtual array element positions are taken as the measurement data set, and the fragment array data of the interrupted array fragment with 4 array elements is taken as the verification data set. At this time, the target azimuth is 10° and 12°, the signal-to-noise ratio is -10dB, and the number of snapshots is 10. It can be seen from the figure that the method proposed in the present invention can achieve the correct azimuth estimation of the target when the virtual array is discontinuous due to expansion.

[0118] Simulation Experiment 2

[0119] like Figure 4 and Figure 5 As shown in , another possible reason for the virtual array to have "holes" and generate array fragments is that there are elements in the physical array that fail. Take the coprime array parameters M = 2, N = 3, and the array expansion factor β = 2. At this time, the coprime array does not have virtual array "holes", and the virtual array obtained after expansion is continuous. On this basis, let the physical elements 2 and 6 be the elements with abnormal working status. When the elements 2 and 6 fail, the virtual array obtained at this time will have "holes", as shown in Figure 4At this time, the target orientations are 13° and 18°, the signal-to-noise ratio is 10dB, the number of snapshots is 10, and the orientation results obtained based on the present invention are as follows: Figure 5 As shown in the figure, it can be seen that in this case, the present invention can achieve the correct estimation of the target signal direction.

[0120] Simulation Experiment 3

[0121] like Figure 6 As shown in the figure, the experimental simulation analyzes the mean square error (MSE) of the non-continuous virtual array signal sparse reconstruction method based on virtual array fragment expansion under different signal-to-noise ratios and signal snapshot numbers:

[0122]

[0123] in represents the value of θ in the i-th independent trial k DOA estimation results, I and K represent the number of Monte Carlo experiments and the number of signal sources, respectively.

[0124] In the simulation, M=2, N=3, array expansion factor β=6, and target azimuths are 20° and 23° respectively. Let array elements 3, 4, 22, and 23 be invalid array elements, and the verification data is the array fragment containing 5 virtual array elements obtained after the virtual array is interrupted by a "hole". The number of signal snapshots is 10. When the signal-to-noise ratio changes, the statistical results of MSE after 1000 independent tests are obtained as follows: Figure 6 As shown in (a). As can be seen from the figure, when the signal-to-noise ratio gradually increases, the MSE decreases and tends to be flat; when the signal-to-noise ratio is fixed at -10dB, other simulation conditions remain unchanged, the MSE changes with the number of signal snapshots as shown in Figure 6 As shown in (b), it can be seen from the figure that when the number of signal snapshots is greater than 15, the MSE gradually tends to be constant, while the MSE obtained by other methods also slowly decreases as the signal-to-noise ratio and the number of snapshots gradually increase, but overall, the MSE is higher.

[0125] Simulation experiment 4

[0126] like Figure 7 and Figure 8 As shown in Figure 2, this part of the simulation verifies the performance of the present invention when there is an error in the position of the array element. Considering the robustness of the algorithm, the method of the present invention is compared with the highly robust physical array element domain algorithm. Assuming that there is a position error of 15% of the unit spacing between the array elements, the array element position is as follows: Figure 7As shown in the figure. The simulation conditions are M=2, N=3, array expansion factor β=6, and the target azimuths are 20° and 23° respectively. Let the 3rd, 4th, 22nd and 23rd array elements be invalid array elements. The verification data is the fragment array containing 5 virtual array elements obtained after the virtual array is interrupted by the "hole". The number of signal snapshots is 100, and the signal-to-noise ratio is -5dB. When the target azimuths are 20° and 24° respectively, as shown in the figure, Figure 8 As shown in (a), it can be seen from the figure that the robustness of the algorithm to the array position error is the same as that of the algorithm based on the physical array element domain, and both can stably obtain the target azimuth information. When the target azimuth is close, the azimuths are 20° and 23° respectively, as shown in Figure 8 As shown in (b), at this time, the method based on the physical array element domain can no longer distinguish the directions of the two targets, while the method of the present invention can still achieve correct target direction estimation.

[0127] The present invention also discloses an array signal sparse reconstruction device based on virtual array segment expansion, which is used to implement the above method, including:

[0128] A parameter acquisition unit, used to acquire hardware parameters of the initial array, the hardware parameters including the total number of array elements and the array aperture; acquire configuration parameters of a special array-coprime array according to the hardware parameters, the configuration parameters including an array expansion cardinality (M, N) and an array expansion factor β of two sub-arrays;

[0129] An extended coprime physical array construction unit is used to obtain array element requirements, and expand the coprime array in the physical array element domain based on the array element requirements and configuration parameters, thereby constructing an extended coprime physical array;

[0130] A signal construction unit, used for constructing a far-field signal of an extended mutually prime physical array;

[0131] An extended coprime array virtual array construction unit is used to model the extended coprime array receiving signal, obtain the steering matrix corresponding to each virtual array element in the virtual array, and construct an extended coprime array virtual array based on parameter configuration based on the steering matrix corresponding to each virtual array element;

[0132] A judging unit, used for judging the continuity of the extended coprime array virtual array;

[0133] A first target estimation unit, used for estimating the target position of a uniform linear array based on a continuous virtual array;

[0134] The second target estimation unit is used to segment the virtual array, select the two longest segments in the virtual array fragments, perform signal modeling based on the longer segment of the selected array, select another segment of the array signal as a verification array, and realize non-continuous virtual array signal reconstruction based on the virtual array fragments, and finally obtain the virtual array information and high-resolution target azimuth estimation result output.

[0135] For the array signal sparse reconstruction device based on virtual array fragment expansion of the present invention, since it corresponds to the array signal sparse reconstruction method based on virtual array fragment expansion in the above embodiment, the description is relatively simple. For relevant similarities, please refer to the description of the array signal sparse reconstruction method based on virtual array fragment expansion in the above embodiment, which will not be described in detail here.

[0136] The present invention also discloses a storage medium, which includes a stored program, wherein when the program is run, the method described in any one of the above claims is executed.

[0137] The present invention also discloses an electronic device, comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the processor executes the method described in any one of the above claims through the operation of the computer program.

[0138] In the above embodiments of the present invention, the description of each embodiment has its own emphasis. For parts that are not described in detail in a certain embodiment, reference can be made to the relevant descriptions of other embodiments.

[0139] In the several embodiments provided in this application, it should be understood that the disclosed technical content can be implemented in other ways. Among them, the device embodiments described above are only schematic. For example, the division of the units can be a logical function division. There may be other division methods in actual implementation. For example, multiple units or components can be combined or integrated into another system, or some features can be ignored or not executed. Another point is that the mutual coupling or direct coupling or communication connection shown or discussed can be through some interfaces, indirect coupling or communication connection of units or modules, which can be electrical or other forms.

[0140] The units described as separate components may or may not be physically separated, and the components shown as units may or may not be physical units, that is, they may be located in one place or distributed on multiple units. Some or all of the units may be selected according to actual needs to achieve the purpose of the present embodiment.

[0141] In addition, each functional unit in each embodiment of the present invention may be integrated into one processing unit, or each unit may exist physically separately, or two or more units may be integrated into one unit. The above-mentioned integrated unit may be implemented in the form of hardware or in the form of software functional units.

[0142] If the integrated unit is implemented in the form of a software functional unit and sold or used as an independent product, it can be stored in a computer-readable storage medium. Based on this understanding, the technical solution of the present invention, in essence, or the part that contributes to the prior art, or all or part of the technical solution can be embodied in the form of a software product, and the computer software product is stored in a storage medium, including a number of instructions for a computer device (which can be a personal computer, a server or a network device, etc.) to perform all or part of the steps of the method described in each embodiment of the present invention. The aforementioned storage medium includes: U disk, read-only memory (ROM, Read-Only Memory), random access memory (RAM, Random Access Memory), mobile hard disk, magnetic disk or optical disk and other media that can store program codes.

[0143] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention, rather than to limit it. Although the present invention has been described in detail with reference to the aforementioned embodiments, those skilled in the art should understand that they can still modify the technical solutions described in the aforementioned embodiments, or replace some or all of the technical features therein with equivalents. However, these modifications or replacements do not cause the essence of the corresponding technical solutions to deviate from the scope of the technical solutions of the embodiments of the present invention.

Claims

1. A method for sparse reconstruction of array signals based on virtual array fragment expansion, characterized in that: The following steps are involved: S1. Obtaining hardware parameters of the initial array, the hardware parameters including the total number of array elements and the array aperture; obtaining nonlinear array-coprime array configuration parameters according to the hardware parameters, the configuration parameters including array expansion cardinality (M, N) and array expansion factors β of two sub-arrays; S2, obtaining array element requirements, and expanding the coprime array in the physical array element domain based on the array element requirements and configuration parameters, thereby generating an extended coprime physical array; S3, constructing the far-field signal of the extended mutually prime physical array; S4, modeling the extended coprime array receiving signal, obtaining the steering matrix corresponding to each virtual array element in the virtual array, and constructing the extended coprime array virtual array based on parameter configuration based on the steering matrix corresponding to each virtual array element; S5, judging the continuity of the extended coprime array virtual array, if the obtained extended coprime array virtual array is a continuous virtual array, executing S6, otherwise executing S7; S6, target azimuth estimation of uniform linear array based on continuous virtual array; S7. Segment the virtual array, select the two longest segments in the virtual array fragments, perform signal modeling based on the longer segment of the selected array, select another segment of the array signal as a verification array, and realize non-continuous virtual array signal reconstruction based on the virtual array fragments, and finally obtain the virtual array information and high-resolution target azimuth estimation result output.

2. The array signal sparse reconstruction method based on virtual array segment expansion according to claim 1, characterized in that: Constructing the far-field signal of an extended coprime physical array, including: The far-field narrowband signal is constructed as follows: X(t)=[x1(t),x2(t),…x P (t)] T =Az(t)+N(t) where z(t)=[z1(t),z2(t),…z L (t)] T is the target signal vector received by the array, L is the number of far-field narrowband signals, is a statistically independent Gaussian noise component, P is the number of array elements, I is the unit matrix, is the noise power, A=[a(θ0),a(θ1),...,a(θ L )] is the array flow matrix, 1≤l≤L,a(θ l ) is the array flow vector of the target in the array, d2, d3, ...d P is the distance between two adjacent array elements, where d2 is the distance between the first array element and the second array element, d3 is the distance between the second array element and the third array element, and d P is the distance between the P-1th array element and the Pth array element, λ is the signal wavelength, θ l is the signal incident azimuth; When the processed signal is a broadband signal, the far-field broadband signal is constructed according to the following method: The broadband signal is equally divided into several narrowband signals, and then the signal is constructed based on the construction method of the narrowband signal. The pre-examination processing frequency band range of the far-field broadband is [f L ,f H ].

3. The array signal sparse reconstruction method based on virtual array segment expansion according to claim 1, characterized in that: Modeling the extended coprime array received signal includes: Construct the covariance matrix of the received signal of the extended coprime array virtual array based on parameter selection: Where T is the signal length, X(t) is the array receiving signal, and X H (t) is the conjugate transpose of the signal, and the covariance is vectorized to obtain: Among them, A v is the steering matrix corresponding to each virtual array element in the virtual array, u=vec(I), p=[p1,p2,…,p L ] T is the power of the incident signal.

4. The array signal sparse reconstruction method based on virtual array segment expansion according to claim 1, characterized in that: Target azimuth estimation of uniform linear array based on continuous virtual array, including: Using the traditional robust beamforming method, a virtual uniform linear array flow vector is constructed to obtain the virtual array beam output and target azimuth information.

5. The array signal sparse reconstruction method based on virtual array segment expansion according to claim 1, characterized in that: Target azimuth estimation of uniform linear array based on continuous virtual array, including: Using the high-resolution subspace target direction estimation method, a virtual uniform linear array signal covariance is constructed, and target direction estimation is achieved through signal and noise subspace processing.

6. The array signal sparse reconstruction method based on virtual array segment expansion according to claim 1, characterized in that: Target azimuth estimation of uniform linear array based on continuous virtual array, including: Using the principle of compressed sensing, virtual uniform linear array target sparse recovery and target direction estimation are constructed.

7. The array signal sparse reconstruction method based on virtual array segment expansion according to claim 1, characterized in that: Signal modeling is performed based on a longer segment of the selected array, and another segment of the array signal is selected as a verification array, including: The signal vector is constructed as follows: Where V is the measurement vector in the virtual array element domain, V cv is the validation vector: Where p and q represent the element positions of the virtual array, represents the element position in the entire continuous virtual array, D i represents the elements in the array fragment, i is the number of array fragments, let is the iterative result of the received signal covariance quantization, k is the number of iterations, and the sparse estimation result of the kth iteration of sparse reconstruction, then the verification residual can be obtained based on the virtual array fragment of this iteration. When the number of iterations increases, ε k becomes smaller; when the restored signal is overfitted, ε k becomes larger, when ε k The iteration stops when the minimum value is reached: Where E represents the maximum number of iterations.

8. An array signal sparse reconstruction device based on virtual array segment expansion, used to implement the method according to claim 1, characterized in that: include: A parameter acquisition unit is used to acquire hardware parameters of the initial array, wherein the hardware parameters include the total number of array elements and the array aperture; and to acquire nonlinear array coprime array configuration parameters according to the hardware parameters, wherein the configuration parameters include array expansion cardinality (M, N) and array expansion factors β of two sub-arrays. An extended coprime physical array construction unit is used to obtain array element requirements, and expand the coprime array in the physical array element domain based on the array element requirements and configuration parameters, thereby constructing an extended coprime physical array; A signal construction unit, used for constructing a far-field signal of an extended mutually prime physical array; An extended coprime array virtual array construction unit is used to model the extended coprime array receiving signal, obtain the steering matrix corresponding to each virtual array element in the virtual array, and construct an extended coprime array virtual array based on parameter configuration based on the steering matrix corresponding to each virtual array element; A judging unit, used for judging the continuity of the extended coprime array virtual array; A first target estimation unit, used for estimating the target position of a uniform linear array based on a continuous virtual array; The second target estimation unit is used to segment the virtual array, select the two longest segments in the virtual array fragments, perform signal modeling based on the longer segment of the selected array, select another segment of the array signal as a verification array, and realize non-continuous virtual array signal reconstruction based on the virtual array fragments, and finally obtain the virtual array information and high-resolution target azimuth estimation result output.

9. A storage medium, characterized in that: The storage medium includes a stored program, wherein when the program is run, the method according to any one of claims 1 to 7 is executed.

10. An electronic device comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein: The processor executes the method according to any one of claims 1 to 7 by running the computer program.

Images