Underwater target recognition method and positioning method based on rectangular sparse matrix

Abstract

The application belongs to the technical field of underwater acoustic detection, and provides an underwater target identification method and positioning method based on a rectangular sparse array, wherein the rectangular sparse array comprises two linear arrays extending along an X-axis and two linear arrays extending along a Y-axis, the X-axis is perpendicular to the Y-axis, and the underwater target identification method comprises: combining adjacent linear arrays in the rectangular sparse array to obtain at least two L-shaped arrays; performing joint beam forming processing on the received signals of each L-shaped array to obtain a joint energy distribution matrix of each L-shaped array in a two-dimensional angle space; and identifying a real underwater target existing in a detection area based on the distribution of energy peaks in the joint energy distribution matrix of each L-shaped array. The application utilizes the geometric structure of the rectangular sparse array, constructs L-shaped arrays with different references, and effectively solves the false peak problem of a single L-shaped array in multi-target identification by performing joint processing on the beam forming results of different L-shaped arrays.

Description

Technical Field

[0001] This application belongs to the field of underwater acoustic detection technology, and relates to three-dimensional sonar beamforming technology. Specifically, it provides an underwater target identification and positioning method based on a rectangular sparse array. Background Technology

[0002] Three-dimensional sonar can acquire three-dimensional spatial information of underwater targets and has important application value in underwater detection, marine engineering, target identification and other fields. In order to achieve high angular resolution and low sidelobe level, conventional three-dimensional sonar usually uses a rectangular planar array to receive echo signals, which requires a large number of transducer array elements. The large number of array elements not only leads to a sharp increase in hardware cost, system power consumption and data transmission volume, but also puts extremely high demands on the computing power of subsequent real-time beamforming processing, thus limiting its application in the field of underwater target detection.

[0003] To reduce the number of array elements, sparse array configurations such as L-shaped linear arrays and cross-shaped linear arrays have been proposed. Taking the L-shaped linear array as an example, it consists of two orthogonal linear arrays. Theoretical analysis shows that when beamforming a single target, the product of the beam patterns of the two linear arrays (or the joint processing of the covariance matrix) can obtain a "pencil-shaped" beam with the same main lobe as a rectangular array of the same aperture, thereby obtaining the target's azimuth and elevation information.

[0004] However, all existing beamforming methods used in sparse arrays suffer from the "pseudo-target" problem in multi-target scenarios. Specifically, when multiple targets exist in the detection area, joint processing of the beamforming results from two linear arrays can lead to the formation of false energy peak intersections, or "pseudo-peaks," at the locations of non-real targets. This problem severely restricts the application of low-cost 3D sonar using sparse arrays in real-world multi-target environments and is one of the core challenges currently facing low-cost 3D sonar technology. Summary of the Invention

[0005] This application provides an underwater target identification method based on a rectangular sparse array through embodiments. The rectangular sparse array includes two linear arrays extending along the X-axis and two linear arrays extending along the Y-axis, wherein the X-axis is perpendicular to the Y-axis. The method includes the following steps: S1, combine adjacent linear arrays in the rectangular sparse array to obtain at least two L-shaped arrays; S2, perform joint beamforming processing on the received signals of each L-shaped array to obtain the joint energy distribution matrix of each L-shaped array in the two-dimensional angle space; S3 identifies real underwater targets in the detection area based on the distribution of energy peaks in the joint energy distribution matrix of each L-shaped array.

[0006] This application also provides an underwater target localization method based on a rectangular sparse array through embodiments, including the following steps: Using the aforementioned underwater target identification method based on rectangular sparse arrays, the azimuth and elevation angles of the real underwater targets in the detection area are identified and determined. The distance to the real underwater target is determined based on the echo arrival time corresponding to the identified real underwater target.

[0007] The underwater target identification method based on rectangular sparse arrays provided in this application creatively proposes a rectangular sparse array configuration by deeply analyzing the pseudo-peak generation mechanism in the L-shaped array joint beamforming results. By using the matrix sparse array to form multiple L-shaped arrays with different relative positional relationships, the pseudo-peaks are actively "driven" to appear at different positions in the two-dimensional angular space to achieve pseudo-peak identification and removal. This completely solves the pseudo-peak problem that has plagued multi-target imaging of L-shaped sparse arrays with minimal hardware cost, and provides a practical and feasible technical approach for developing a new generation of low-cost, miniaturized, and high-performance three-dimensional imaging sonar. Attached Figure Description

[0008] Figure 1 A stereoscopic view of the appearance of a three-dimensional sonar; Figure 2 This is a schematic diagram of a signal receiving method using a rectangular array. Figure 3 This is a schematic diagram of a signal receiving method using an L-shaped array; Figure 4 A schematic diagram showing the locations of multiple targets in the detection area; Figure 5 This is a schematic diagram of the joint energy distribution matrix obtained by performing joint beamforming processing on the echo signal of only one L-shaped array in a multi-target scenario. Figure 6 This is a schematic diagram of receiving signals using a rectangular sparse array according to an embodiment of this application; Figure 7 This is a schematic diagram of receiving signals using a rectangular sparse array according to another embodiment of this application; Figure 8 A front view of a rectangular sparse array and a sound source provided according to some embodiments of this application; Figure 9 This is a flowchart of an underwater target recognition method based on a rectangular sparse array provided according to an embodiment of this application; Figure 10 This is a schematic diagram of the joint energy distribution matrix obtained by performing joint beamforming processing on the received signal of the first L-shaped array in some embodiments; Figure 11This is a schematic diagram of the joint energy distribution matrix obtained by performing joint beamforming processing on the received signal of the second L-shaped array in some embodiments; Figure 12 This is a schematic diagram of the joint energy distribution matrix obtained by performing joint beamforming processing on the received signal of the third L-shaped array in some embodiments. Figure 13 This is a schematic diagram of the joint energy distribution matrix obtained by performing joint beamforming processing on the received signal of the fourth L-shaped array in some embodiments. Figure 14 This is a schematic diagram of the first L-shaped array obtained by combining elements in some embodiments; Figure 15 This is a schematic diagram of the second L-shaped array obtained by combining elements in some embodiments; Figure 16 This is a schematic diagram of the third L-shaped array obtained by combining elements in some embodiments; Figure 17 This is a schematic diagram of the fourth L-shaped array obtained by combining elements in some embodiments; Figure 18 This is a schematic diagram of the time-domain signal received by the leftmost element of the second linear array in a specific embodiment; Figure 19 This is a schematic diagram of the time-domain signal received by the rightmost element of the second linear array in a specific embodiment; Figure 20 This is a schematic diagram of the time-domain signal that a virtual array element at the center of a rectangular sparse array is expected to receive in a specific embodiment. Figure 21 This is a schematic diagram of a three-dimensional sonar image obtained by an underwater target localization method based on a rectangular sparse array according to some embodiments of this application. Detailed Implementation

[0009] The present application will now be further described based on preferred embodiments and with reference to the accompanying drawings.

[0010] In the description of the embodiments of this application, it should be noted that the terms "upper," "lower," "inner," and "outer," etc., indicating orientation or positional relationships based on the orientation or positional relationships shown in the accompanying drawings, or the orientation or positional relationships commonly used when the product of this application is in use, are only for the convenience of describing this application and simplifying the description, and do not indicate or imply that the device or element referred to must have a specific orientation, or be constructed and operated in a specific orientation, and therefore should not be construed as a limitation on this application. Furthermore, in the description of this application, the terms "first," "second," etc., are used to distinguish different units, but these are not limited by the manufacturing order, nor should they be construed as indicating or implying relative importance. Their names may differ in the detailed description and claims of this application. In addition, for ease of understanding, various components in the drawings have been enlarged or reduced, but this is not intended to limit the scope of protection of this application.

[0011] The vocabulary used in this specification is for illustrative purposes and is not intended to limit the scope of this application. It should also be noted that, unless otherwise expressly specified and limited, the terms "set," "connected," and "linked" should be interpreted broadly. For example, they can refer to a fixed connection, a detachable connection, or an integral connection; they can refer to a mechanical connection, a direct connection, or an indirect connection via an intermediate medium; or they can refer to the internal communication between two components. Those skilled in the art will understand the specific meaning of these terms in this application.

[0012] Figure 1 A physical image of a three-dimensional sonar used for underwater target detection, such as... Figure 1 As shown, the core functional module of this three-dimensional sonar is a sound source and receiving unit. The sound source and receiving unit are fixedly connected to the shell. The sound source is preferably a directional sound source. Based on the received control command (the control command can be transmitted to the sound source through a watertight signal cable, or issued through a control unit set in a watertight manner inside the shell), it emits narrowband or broadband pulse signals to the detection area according to a preset pulse interval. When an underwater target is present in the detection area, the emitted pulse signal will be reflected on the surface of the underwater target, thereby generating an echo signal.

[0013] The receiving unit consists of several transducer array elements, as shown in the figure. Its surface is coated with waterproof coating (black part) and other materials. Each transducer array element is used to receive underwater acoustic signals. Specifically, after the sound source transmits a pulse signal each time, each array element performs signal acquisition within a preset time window to obtain the received signal corresponding to each transmitted pulse signal. The starting position of the time window has a sufficient delay relative to the time of transmitting the pulse signal to avoid receiving a direct wave.

[0014] The received signals of each array element can be received and processed by a signal processing unit located inside the hull, or they can be transmitted to a ship-based or shore-based signal processing unit for signal processing via a communication module and communication cable.

[0015] The signal processing unit performs beamforming on the received signals from multiple transducer array elements to scan the detection area in all directions. When the beamforming direction is aligned with the return direction of the echo signal, the signal energy obtained by beamforming will be significantly enhanced, thereby enabling the direction identification of underwater targets. Combined with the round-trip time of the echo signal, the three-dimensional positioning or imaging of underwater targets can be achieved.

[0016] Furthermore, it is understood that the three-dimensional sonar can be powered either by a power supply module such as a battery installed inside the shell, or by drawing power from the ship base or the ship base's power source using a power cable.

[0017] Based on the number and arrangement of transducer array elements contained in the receiving unit, the receiving unit can be divided into planar array configuration and sparse linear array configuration. Figure 2 A rectangular array configuration is shown, in which multiple transducer elements are arranged at equal intervals along mutually perpendicular X-axis and Y-axis directions, thereby forming a matrix on the XY plane. Without loss of generality, the array is... Indicates the number of array elements along the X-axis direction, in... Let represent the number of array element rows along the Y-axis. Then, the total number of array elements in the entire rectangular array is: ,For example, Figure 2 In the embodiment shown, That is, a total of including Each array element.

[0018] Using a planar array configuration for underwater acoustic signal reception allows for direct reception at the azimuth angle. and pitch angle Beamforming scanning operations are performed in the constructed two-dimensional directional space, with reference to... Figure 2 For any incident sound ray, the azimuth angle The pitch angle refers to the angle between the projection of the incident sound ray onto the XZ plane and the YZ plane. This refers to the angle between the incident sound ray and the XZ plane, where the Z-axis is perpendicular to the XY plane, meaning the Z-axis direction is the normal direction of the plane containing the transducer element. Figure 2 Taking underwater target 1 as an example, under far-field conditions, the line connecting its center and the center of the array is taken as the incident direction of the echo signal after the underwater acoustic signal emitted by the sound source is reflected. Then, by analyzing the received signal of the rectangular array (i.e., the received signal of each transducer element)... Two-dimensional beamforming operations in space can determine the azimuth and elevation angles of underwater target 1. and , and then combined with The round-trip time of the peak value of the underwater acoustic signal received in the direction can be used to determine the distance of underwater target 1 from the center of the rectangular array. .

[0019] This can be directly achieved using a rectangular array configuration. Two-dimensional beamforming is possible, but to achieve higher angular resolution and lower sidelobe level, it is necessary to increase the... , The quantity, for example, when the sound source frequency reaches 500kHz, in order to ensure resolution, and It can reach 100 or even more, causing the total number of array elements to reach This high-density array element deployment inevitably leads to a sharp increase in hardware costs, system power consumption, and data transmission volume, and also places extremely high demands on the computing power of subsequent real-time beamforming processing.

[0020] To reduce the number of array elements, sparse array configurations such as L-shaped arrays and cross-shaped arrays have been proposed. Figure 3 An L-shaped array configuration is shown, consisting of two linear arrays orthogonal at their endpoints. The number of elements in this L-shaped array is... While keeping overall dimensions and aperture parameters constant, the total number of array elements is much smaller than that of a rectangular array. Furthermore, theoretical analysis shows that by beamforming the received signals from two of the orthogonal linear arrays separately, two... By generating a two-dimensional energy map in space and then performing element-wise multiplication of the two two-dimensional intensity maps, a "pencil-shaped" beam with the same main lobe as a rectangular array of the same aperture can be obtained. This directional beam can then be used to scan at various azimuth and elevation angles, enabling the identification of underwater target directions. Correspondingly, this method utilizes two orthogonal linear arrays for... The operation of two-dimensional beamforming in space can be called joint beamforming.

[0021] However, since each of the two linear arrays only possesses high resolution within a single angular range, such as Figure 3 The linear array along the X-axis only has azimuth angle. The high resolution, the linear array along the Y-axis only has elevation angle. The high resolution means that when the two linear arrays' two-dimensional energy maps are multiplied point-by-point, a "wake" will appear at the intersection with the real target, especially when the detection area is such that... Figure 4 When multiple underwater targets are present, the wakes of different targets intersect and intensify, as shown in the diagram. Figure 5 As shown, in the two-dimensional energy map obtained by joint beamforming, in addition to the locations of multiple underwater targets... The actual target peak will appear at the coordinates, and the wakes of various underwater targets will also intersect and reinforce each other, resulting in multiple false "peaks" in the image.

[0022] Clearly, the appearance of the aforementioned spurious peaks is determined by the mechanism of two-dimensional beamforming using a sparse linear array configuration. It is difficult to identify or eliminate them by means of feature recognition and noise reduction alone. Targeted treatment must be carried out to address the root cause of the problem. To this end, this application proposes an underwater target identification method based on a rectangular sparse array. This method processes the received signal of a rectangular sparse array composed of four linear arrays, ensuring that the real target can be accurately identified regardless of whether there is a single underwater target or multiple underwater targets in the detection area. At the same time, it effectively eliminates false targets corresponding to spurious peaks in the joint beamforming results.

[0023] Figure 6 The figure shows a schematic diagram of receiving underwater acoustic signals using a rectangular sparse array as a receiving unit in one embodiment. As shown, the rectangular sparse array includes two linear arrays extending along the X-axis and two linear arrays extending along the Y-axis (as mentioned above, the X-axis is perpendicular to the Y-axis). Obviously, these four linear arrays are in the same XY plane, and the direction they point to the detection area is the normal direction of the XY plane, that is, the Z-axis direction.

[0024] Furthermore, each linear array is composed of multiple array elements arranged at intervals. As mentioned above, each array element collects the received signal within a preset time window after the sound source emits a pulse signal.

[0025] Figure 7 A schematic diagram of receiving underwater acoustic signals via a rectangular sparse array is shown in another embodiment. Figure 8 It shows the relationship with Figure 7 The front view corresponding to the embodiment shown. Figure 7 , Figure 8 The illustrated embodiments and Figure 6 The difference in the embodiments lies in the different placement of the sound source; specifically, in Figure 6 In the embodiment shown, the sound source is located outside the rectangular sparse array, i.e., using a method similar to... Figure 1 A similar setup method, while Figure 7 , Figure 8 In the illustrated embodiment, the sound source is located inside the rectangular sparse array, preferably at the center of the rectangle.

[0026] pass Figures 6 to 8 As can be seen from the embodiments shown, the method provided in this application does not have any special limitations on the position of the sound source relative to the rectangular sparse array, as long as the sound source is oriented towards the detection area, that is, emitting underwater acoustic signals in the Z-axis direction in the figure.

[0027] In this application, for the convenience of describing the linear arrays constituting the rectangular sparse matrix, reference is made to... Figure 8 Facing the rectangular sparse array from the detection area, in a counterclockwise direction, the linear array extending along the Y-axis on the left is called the first linear array L1, the linear array extending along the X-axis at the bottom is called the second linear array L2, the linear array extending along the Y-axis on the right is called the third linear array L3, and the linear array extending along the X-axis at the top is called the fourth linear array L4.

[0028] Accordingly, the first linear array L1 and the third linear array L3 each contain The second linear array L2 and the fourth linear array L4 each contain array elements arranged at equal intervals. The rectangular sparse array contains 1,000 array elements arranged at equal intervals. The first linear array L1 and the second linear array L2 share the same array element at the lower left corner of the rectangle; the second linear array L2 and the third linear array L3 share the same array element at the lower right corner of the rectangle; the third linear array L3 and the fourth linear array L4 share the same array element at the upper right corner of the rectangle; and the fourth linear array L4 and the first linear array L1 share the same array element at the upper left corner of the rectangle. Therefore, the total number of array elements in the entire rectangular sparse array is 1. (like Figures 6 to 8 In the embodiment, the number of elements in the rectangular sparse array is The number of elements in a single L-shaped array is on the same order of magnitude as that in a rectangular array, and is also an order of magnitude smaller than that in a rectangular array.

[0029] In some specific embodiments, the elements of the rectangular sparse array can be arranged on the support material inside the three-dimensional sonar shell according to the above configuration and watertight. The arrangement method is the same as that of existing three-dimensional sonars using area arrays, and will not be described again here.

[0030] Understandable, Figures 6 to 8 This diagram is only intended to illustrate the construction of a rectangular sparse array. The number of array elements, spacing, and fixing method of each linear array in the diagram do not constitute a limitation on the configuration of this rectangular sparse array. Those skilled in the art can determine the appropriate number and size of array elements based on specific underwater target identification indicators, such as detection distance and accuracy, combined with parameters such as sound source frequency.

[0031] Figure 9 A flowchart of an underwater target recognition method based on a rectangular sparse array according to an embodiment of this application is shown, as follows: Figure 9 As shown, the underwater target identification method includes the following steps: S1, combine adjacent linear arrays in the rectangular sparse array to obtain at least two L-shaped arrays; S2, perform joint beamforming processing on the received signals of each L-shaped array to obtain the joint energy distribution matrix of each L-shaped array in the two-dimensional angle space; S3 identifies real underwater targets in the detection area based on the distribution of energy peaks in the joint energy distribution matrix of each L-shaped array.

[0032] The specific implementation methods of steps S1 to S3 described above will be explained in detail below with reference to the accompanying drawings.

[0033] In step S1, any two adjacent linear arrays can be extracted from the rectangular sparse matrix and combined. Since any two adjacent linear arrays are perpendicular to each other and share array elements, an L-shaped matrix is ​​obtained after combination. Obviously, for a rectangular sparse matrix, there are four ways to combine them. Without loss of generality, we can... Figure 8 The combination of the first linear array L1 and the second linear array L2 is called the first L-shaped array; the combination of the second linear array L2 and the third linear array L3 is called the second L-shaped array; the combination of the third linear array L3 and the linear array L4 is called the third L-shaped array; and the combination of the fourth linear array L4 and the first linear array L1 is called the fourth L-shaped array. Therefore, the number of L-shaped arrays obtained in step S1 can be two, three, or four.

[0034] In step S2, the received signals of each L-shaped array obtained by combination, that is, the received signals of all array elements contained in each L-shaped array, are subjected to joint beamforming processing to obtain the two-dimensional beam energy distribution matrix corresponding to each L-shaped array.

[0035] Specifically, in the embodiments of this application, the underwater acoustic signal emitted by the sound source toward the detection area is a pulse signal, that is, a narrowband or broadband signal is sent at a preset time interval. After the transmission of each pulse signal is completed, each array element receives the signal simultaneously within a preset time window to obtain the received signal corresponding to each pulse signal. The position and duration of the time window can be set according to the detection distance index. When there is an underwater target in the detection area, the pulse signal emitted by the sound source will be reflected at the underwater target. After the reflected signal is received by each array element, a corresponding echo is generated at the position corresponding to the arrival time in the received signal of each array element.

[0036] Furthermore, it is understandable that, apart from the echoes formed by reflections from underwater targets, in the absence of other interfering sound sources in the detection area, the received signals of each array element at other times are underwater environmental noise signals.

[0037] In some optional embodiments, joint beamforming processing is performed on the received signal of any L-shaped array, including the following steps: S21, beamforming processing is performed independently on the received signals of the two adjacent linear arrays that make up the L-shaped array, so as to obtain the energy distribution matrix of each linear array in the two-dimensional angular space.

[0038] The specific implementation of beamforming for each linear array is known to those skilled in the art. It is understood that, since each L-shaped array includes a linear array extending along the X-axis and a linear array extending along the Y-axis, without loss of generality, the linear array extending along the X-axis in an L-shaped array can be called a horizontal linear array, and the linear array extending along the Y-axis can be called a vertical linear array. The horizontal and vertical linear arrays are then numbered in two dimensions.

[0039] Specifically: For a horizontal linear array within an L-shaped array, all its elements have the same Y-axis coordinate, meaning each element can be numbered as follows: ,in, This represents the total number of elements in a horizontal linear array. Let the row numbering of the elements shared by the horizontal and vertical linear arrays in the vertical linear array be used. Taking the first L-shaped array composed of the first linear array L1 and the second linear array L2 as an example, its horizontal linear array is the second linear array L2. Therefore, in the first L-shaped array, the row numbering of each element in the horizontal linear array is... This refers to the row number of the bottom element in the vertical linear array (first linear array L1).

[0040] For a vertical linear array in an L-shaped array, all its elements have the same X-axis coordinate, meaning that each element can be numbered as follows: ,in, This represents the total number of elements in a vertical linear array. To assign column numbers to the elements shared by the vertical and horizontal linear arrays within the horizontal linear array, taking the first L-shaped array as an example, its vertical linear array is the first linear array. Therefore, in the first L-shaped array, the column numbers of each element in the vertical linear array... This refers to the column number of the leftmost element in the horizontal linear array (second linear array L2).

[0041] Clearly, the same principle applies to the second, third, and fourth L-shaped arrays, which can also be determined based on the row and column numbers of the shared array elements. and The specific value.

[0042] For any L-shaped array, beamforming can be performed independently on its horizontal and vertical linear arrays using the following formula: (1), in, The beamforming processing result for a horizontal linear array. The result of beamforming processing for a vertical linear array. Numbering is done by discrete distance. , The discrete azimuth and discrete elevation angles are numbered respectively. For the horizontal linear array The original received signal of each array element For the vertical linear array The original received signal of each array element For the horizontal linear array The coordinates of each array element in the XY plane, where the origin of the XY plane is the center of the rectangular sparse matrix. To direct the shaped beam of the horizontal linear array And focus on And it is necessary to address the first The delay applied to the original received signal of each array element, For the vertical linear array The coordinates of each array element in the XY plane To direct the beam of the vertical linear array And focus on And it is necessary to address the first The delay applied to the original received signal of each array element.

[0043] The general expression is: (2), in, This is the speed of sound in water.

[0044] Under far-field conditions, the delay expression can be approximated by the following formula: (3).

[0045] When beamforming the horizontal and vertical linear arrays respectively, first substitute the coordinate positions of the corresponding array element numbers into equation (2) to obtain the delay that needs to be applied to the original received signal of each array element. Then substitute it into equation (1) to obtain the delay in each discrete azimuth-elevation direction, i.e., in the two-dimensional angle space. place , Considering that for a given distance... Only the received signal needs to be processed. The corresponding time-domain signal, i.e., focused on distance. Therefore , All are two-dimensional angular spaces ( The energy distribution matrix in space, without needing to be explicitly shown. The first in the matrix The values ​​of each element represent the values ​​in the following elements: The signal energy received in the direction of travel.

[0046] It is understood that, in the embodiments of this application, the energy distribution should be interpreted broadly, and it can be either... The sound pressure level at a given location can also be called sound intensity; its specific dimensions depend on the processing performed on the original received signal. The form only needs to ensure The magnitude of the signal should be sufficient to reflect its strength.

[0047] S22, perform a multiplication operation on the elements at the same position in the energy distribution matrices of the two adjacent linear arrays obtained in step S21 to obtain the joint energy distribution matrix of the L-shaped array in two-dimensional angular space.

[0048] In some embodiments, the joint energy distribution matrix of the L-shaped array can be obtained using the following formula. : (4).

[0049] To maintain dimensional consistency, in some preferred embodiments, the right-hand side of equation (4) can be squared to achieve dimensional uniformity, and then normalization can be performed to obtain the normalized joint energy distribution matrix: (5).

[0050] In addition, it can be noted For a specific distance A function describing the two-dimensional angular spatial energy distribution at a given distance, i.e., focused at a specific distance. For different Each can obtain a In some alternative embodiments, all focusing distances can also be [discussed / captured]. of By performing element-wise accumulation, a joint energy distribution matrix is ​​obtained that is independent of the distance to the underwater target and only depends on the two-dimensional angle.

[0051] Figures 10 to 13 The diagrams illustrate, in a specific embodiment, the normalized joint energy distribution matrix obtained by performing joint beamforming operations on the first, second, third, and fourth L-shaped arrays. In this embodiment, the detection area includes three equidistant underwater targets. The coordinates are: ( ), ( ), ( ).

[0052] exist Figures 10 to 13In the illustrated embodiment, the sound source emits a narrowband pulse signal with a center frequency of 500 kHz, a bandwidth of 200 Hz, and a pulse time interval of 300 ms, corresponding to a detection distance of approximately 200 meters. Each linear array constituting the rectangular sparse array consists of 100 transducer elements arranged at equal intervals of half wavelength (1.5 mm). The length of each linear array is 14.85 cm, and the main lobe beamwidth in the normal direction is 1.05°. The total number of elements in the entire rectangular sparse array is 396, which is far less than the total number of elements in a surface array of the same size (100 × 100 = 10,000).

[0053] It is understood that the parameters of the aforementioned sound source and receiving array do not constitute a limitation on the technical solution of this application. Those skilled in the art can adjust one or more of the above parameters according to the specific underwater target identification scenario and detection accuracy indicators. For example, in order to improve... The resolution of the coordinates can be increased by increasing the number of elements per linear array to improve the aperture (depending on the total number of elements and the spacing between them), or by adjusting the bandwidth of the pulse signal emitted by the sound source (from hundreds of Hz to several kHz) to adjust the aperture. Effective detection range and detection accuracy of coordinates.

[0054] As can be seen, in the joint energy distribution matrix obtained by processing data received from different L-shaped arrays, the energy peaks of the three real underwater targets are in two dimensions. While the coordinate values ​​in angular space remain unchanged, the "pseudo-peaks" formed by the wake intersections appear at different positions in different joint energy distribution matrices. The following combines... Figures 14 to 20 The above phenomena will be analyzed.

[0055] Figures 14 to 17 The diagrams illustrate the positional relationships between multiple underwater targets and the first, second, third, and fourth L-shaped arrays when multiple underwater targets are present in the detection area. It can be seen that because the first linear array L1 and the third linear array L3 are located to the left and right of the matrix center, while the second linear array L2 and the fourth linear array L4 are located to the top and bottom of the matrix center, although two different L-shaped arrays may share the same linear array, the positional relationship of the other linear array, which differentiates their relative positions, determines the different locations where false peaks are generated.

[0056] Taking the first L-shaped array and the second L-shaped array as examples, they share a horizontal linear array, namely the second linear array L2, but their vertical linear arrays are the first linear array L1 and the third linear array L3, respectively. Figure 18 and Figure 19 The images show the echo signals from the same underwater target received by the leftmost element of the second linear array L2 (which is also the bottommost element of the first linear array L1) and the rightmost element of the second linear array L2 (which is also the bottommost element of the third linear array L3). Figure 20This shows the expected time-domain signal received when a virtual array element is located at the center of a rectangular virtual array, which provides a reference for beamforming of each linear array.

[0057] Combination Figures 18 to 20 It can be observed that when different vertical linear arrays are used for processing with the same horizontal linear array, the received signal itself has a delay. At the same time, due to the different positions relative to the reference, the pseudo-peaks (-10°, -10°) generated by the intersection of multiple real underwater target wakes in the processing result of the first L-shaped array disappear in the processing result of the second L-shaped array, and new pseudo-peaks are generated at other non-overlapping positions (10°, -5°) and (10°, -10°).

[0058] Based on the same principle, the energy peak positions of real underwater targets in the joint energy distribution matrix of the third and fourth L-shaped arrays are the same as those in the first and second L-shaped arrays, while the positions of false peaks are different. Therefore, in step S3, the above characteristics can be used to identify real underwater targets and eliminate false underwater targets by utilizing the different distribution characteristics of energy peaks in the joint energy distribution matrix of at least two L-shaped arrays.

[0059] In some optional embodiments, step S3 includes the following steps: Step S31: Extract energy peaks from the joint energy distribution matrix of each L-shaped array obtained in step S2 to obtain the candidate target set corresponding to each L-shaped array.

[0060] Step S32: Identify the candidate targets that exist in each candidate target set as real underwater targets.

[0061] In some optional embodiments, the number of L-shaped arrays combined in step S1 is two. In step S3, the identification of real underwater targets can be achieved based on the joint energy distribution matrix of the two L-shaped arrays. For example, in Figures 10 to 13 In the illustrated embodiment, the combined L-shaped array is a first L-shaped array and a second L-shaped array. Energy peaks can be extracted from the joint energy distribution matrix of the first and second L-shaped arrays according to a preset energy threshold. The magnitude of the energy threshold should ensure accurate differentiation between the energy peak and horizontally or vertically extending tails. For example, in some optional embodiments, when each element in the joint energy distribution matrix is ​​represented using a normalized decibel value, the energy threshold is -3dB. Elements greater than this threshold correspond to two-dimensional angular coordinates. It can be identified as a candidate target, indicating that there may be an underwater target in that direction.

[0062] To each Figure 10 and Figure 11Performing the above operations yields the first candidate target set A1 = {(-10°, -5°), (-5°, -10°), (-10°, -10°), (10°, 20°)} corresponding to the first L-shaped array, and the second candidate target set A2 = {(-10°, -5°), (-5°, -10°), (10°, -5°), (10°, -10°), (10°, 20°)} corresponding to the second L-shaped array. The candidate targets simultaneously present in A1 and A2 are (-10°, -5°), (-5°, -10°), and (10°, 20°). Therefore, it can be determined that in this embodiment, there are three real underwater targets in the detection area, whose two-dimensional angular coordinates relative to the plane of the rectangular sparse array are... They are (-10°, -5°), (-5°, -10°), and (10°, 20°), respectively.

[0063] Accordingly, only in Figure 10 The energy peaks that exist in the middle (-10°, -10°) and only in Figure 11 The energy peaks present at (10°, -5°) and (10°, -10°) can be identified as false peaks caused by joint beamforming, and the corresponding candidate targets can be identified as false underwater targets.

[0064] It is understandable that due to factors such as installation errors of individual elements in different linear arrays, the two-dimensional angular coordinates of the same underwater target may deviate in different joint energy distribution matrices. Therefore, in some preferred embodiments, a two-dimensional angular distance threshold E_M can be set. Candidate targets that exist in different joint energy distribution matrices and whose two-dimensional angular coordinate distances do not exceed E_M can be considered as the same candidate target. The value of E_M can be determined based on the main lobe beamwidth of the linear array constituting the rectangular sparse array. For example, preferably, E_M is set to 1 / 4 of the main lobe beamwidth.

[0065] It is understood that the method provided in this application is not limited to scenarios where there are multiple underwater targets. For example, when there is only one underwater target in the detection area, there are no false peaks in the joint energy distribution matrix of each L-shaped array. In this case, the unique true peak can still be accurately extracted through step S3, thereby identifying the unique true underwater target.

[0066] In some alternative embodiments, step S1 can also yield two L-shaped arrays in other combinations, such as a first L-shaped array and a third L-shaped array, or a third L-shaped array and a fourth L-shaped array. Correspondingly, in step S3, the joint energy distribution matrix of the first L-shaped array and the third L-shaped array (e.g., ...) can be used as a reference. Figure 10 and Figure 12 Underwater target identification can be performed based on the joint energy distribution matrix of the third and fourth L-shaped arrays (as shown). Figure 12 and Figure 13 (As shown) underwater target identification.

[0067] In some optional embodiments, the number of L-shaped arrays obtained by combination can be increased in step S1, and the number of L-shaped arrays used for underwater target identification operations can be increased in S3. For example, in step S1, a first L-shaped array, a second L-shaped array, and a third L-shaped array (or a combination of three other L-shaped arrays) can be obtained by combination. Correspondingly, in step S31, the energy peaks of the joint energy distribution matrix of the first L-shaped array, the second L-shaped array, and the third L-shaped array (or a combination of three other L-shaped arrays) are extracted to obtain three candidate target sets A1, A2, and A3. Then, in step S32, the candidate targets that exist in the three sets at the same time are identified as real underwater targets.

[0068] In some optional embodiments, all four L-shaped arrays, namely the first L-shaped array, the second L-shaped array, the third L-shaped array, and the fourth L-shaped array, can be combined in step S1. Correspondingly, in S31, the energy peaks of the joint energy distribution matrix of all four L-shaped arrays are extracted to obtain all four candidate target sets A1, A2, A3, and A4. Then, in step S33, the candidate targets that exist in all four candidate target sets are identified as real underwater targets.

[0069] Preferably, when all four L-shaped arrays are obtained in step S1, four candidate target sets can be generated accordingly in step S31. At this time, the criteria for judging real underwater targets can be appropriately relaxed. For example, in step S33, a candidate target that exists in any three of the four candidate target sets can be identified as a real underwater target. For example, a candidate target that exists in A1, A2, and A3 but not in A4 can be identified as a real underwater target. By adopting this "majority rule" soft decision method, the system can still robustly output real targets even when a linear array is blocked or loses peak values ​​due to noise interference, thus improving the robustness and adaptability of the system in complex environments.

[0070] This application also provides an underwater target localization method based on a rectangular sparse array, which includes the following steps: Step 1: Using the aforementioned underwater target identification method based on rectangular sparse arrays, identify and determine the azimuth and elevation angles of the actual underwater targets in the detection area; Step 2: Determine the distance to the real underwater target based on the echo arrival time corresponding to the identified real underwater target.

[0071] Specifically, in step one, the aforementioned underwater target recognition method based on a rectangular sparse array is used to identify one or more real underwater targets from the received signals of the rectangular sparse array, and to determine their two-dimensional angular coordinates relative to the rectangular sparse array. Then, it can be extracted from the received signal. The echo signal in the direction is used to determine the total time from signal transmission to reception by using the arrival time of the echo signal, thereby allowing the calculation of the distance of each underwater target relative to the rectangular sparse array. This enables three-dimensional positioning of underwater targets.

[0072] In some alternative embodiments, such as Figure 21 As shown, the matrix element values ​​at the actual location of the underwater target can be set as the echo intensity ( The smaller the value, the greater the strength. The larger the peak, the lower the intensity, thus generating a clean and clear 3D sonar image with false peaks removed.

[0073] The specific embodiments of this application have been described in detail above. For those skilled in the art, several improvements and modifications can be made to this application without departing from the principle of this application, and these improvements and modifications also fall within the protection scope of the claims of this application.

Claims

1. A method for underwater target recognition based on a rectangular sparse array, the rectangular sparse array comprising two linear arrays extending along an X axis and two linear arrays extending along a Y axis, the X axis being perpendicular to the Y axis, characterized in that, Includes the following steps: S1, combine adjacent linear arrays in the rectangular sparse array to obtain at least two L-shaped arrays; S2, perform joint beamforming processing on the received signals of each L-shaped array to obtain the joint energy distribution matrix of each L-shaped array in the two-dimensional angle space; S3, based on the distribution of energy peaks in the joint energy distribution matrix of each L-shaped array, identify the real underwater targets in the detection area; Joint beamforming processing is performed on the received signal of any L-shaped array, including the following steps: S21, beamforming processing is performed independently on the received signals of the two adjacent linear arrays that make up the L-shaped array, so as to obtain the energy distribution matrix of each linear array in the two-dimensional angular space; S22, Multiply the elements at the same position in the energy distribution matrices of the two adjacent linear arrays obtained in step S21 to obtain the joint energy distribution matrix of the L-shaped array in the two-dimensional angular space; Step S3 includes: Step S31: Extract energy peaks from the joint energy distribution matrix of each L-shaped array obtained in step S2 to obtain the candidate target set corresponding to each L-shaped array; Step S32: Identify the candidate targets that exist in each candidate target set as real underwater targets. 2.The method of claim 1, wherein, The two-dimensional angle space is space, wherein, is an azimuth coordinate, is an elevation coordinate.

3. The underwater target identification method based on a rectangular sparse array according to claim 1, characterized in that, Each linear array of the rectangular sparse array is composed of multiple array elements arranged at intervals. Each array element collects the received signal within a preset time window after the sound source emits a pulse signal. The sound source is located inside the rectangular sparse array.

4. The underwater target identification method based on a rectangular sparse array according to claim 1, characterized in that, Candidate targets that exist in different joint energy distribution matrices and whose two-dimensional angular coordinate distances to each other do not exceed E_M are considered as the same candidate target, where E_M is the two-dimensional angular distance threshold.

5. The underwater target identification method based on a rectangular sparse array according to claim 4, characterized in that, E_M is determined based on the main lobe beamwidth of the linear array that constitutes the rectangular sparse array. 6.The method for underwater target recognition based on rectangular sparse matrix according to claim 1, characterized in that, When step S1 combines to obtain four L-shaped matrices... In step S33, candidate targets that exist in any three of the four candidate target sets simultaneously are also identified as real underwater targets.

7. An underwater target positioning method based on a rectangular sparse matrix, characterized in that, Includes the following steps: Using the underwater target identification method based on a rectangular sparse array as described in claim 1, the azimuth and elevation angles of the real underwater targets in the detection area are identified and determined; The distance to the real underwater target is determined based on the echo arrival time corresponding to the identified real underwater target.

Images