A cylindrical array sonar system and a target detection method based on high resolution beamforming

Abstract

The application provides a cylindrical array sonar system and a target detection method based on high-resolution beam forming. The method comprises the following steps: converting a beam forming weight vector of a received sound wave signal into a Kronecker product, combining a joint probability density model to construct a negative log-likelihood function, and optimizing a vector in an azimuth dimension and an elevation dimension through a preset direction constraint condition to form a high-resolution beam; performing correlation analysis on echo signals reflected by a target at different azimuth angles and elevation angles and on transmitted sound wave signals at different time delays; constructing a three-dimensional matrix containing a distance, an azimuth angle and an elevation angle according to a correlation analysis result, performing normalization processing on the three-dimensional matrix to eliminate signal intensity differences, and outputting a target detection positioning result. Through the combination of Kronecker integral decomposition and three-dimensional matrix modeling, the application realizes high-resolution joint positioning detection of a target in an azimuth angle, an elevation angle and a distance dimension, effectively eliminates signal intensity differences, and improves positioning accuracy.

Description

TECHNICAL FIELD

[0001] The present application relates to the field of sonar technology, and in particular to a cylindrical array sonar system and a target detection method based on high-resolution beamforming. BACKGROUND

[0002] Existing active sonar systems are widely used in underwater detection, navigation, imaging and other fields, and usually use array-form sonar transmitting and receiving elements to transmit signals and receive echoes. In order to improve the detection accuracy and coverage, different forms of array configurations are adopted in existing active sonar systems, including linear array, planar array, circular array, etc. Among them, the circular array is more common in application. However, the existing array arrangement has limitations in some cases, especially in target detection accuracy, resolution and target recognition. Traditional sonar systems are usually based on simple array designs such as linear array and rectangular array, which meet the basic detection requirements, but have great challenges in high-precision and high-resolution target detection. Especially in the application scenarios that require accurate positioning or high-resolution imaging, the array configuration of the existing system cannot effectively reduce signal interference and is difficult to achieve precise beam control.

[0003] At the same time, the signal processing method of the existing sonar system is also limited. Although some sonar systems use beamforming technology to enhance signals from a specific direction by adjusting the phase difference of the receiving elements, in complex environments, the existing beamforming algorithm still has problems of insufficient noise suppression, signal distortion and echo interference. These problems result in low accuracy and reliability of target detection, especially in the case of long target distance or large environmental noise.

[0004] In addition, the traditional sonar system usually uses time division mode for signal transmission and reception, which easily leads to low detection efficiency. Although some systems use full-duplex mode (i.e. simultaneous signal transmission and reception), how to effectively avoid signal interference, reduce echo distortion and improve target positioning accuracy is still a difficult problem to be solved. Therefore, with the continuous progress of technology, a new sonar system is needed to effectively improve the detection accuracy and resolution, especially by optimizing the array design and using advanced beamforming technology to improve the target detection performance. SUMMARY

[0005] The present application provides a cylindrical array sonar system and a target detection method based on high-resolution beamforming, to solve the problems of insufficient noise suppression, signal distortion and echo interference in the prior art, as well as low accuracy and reliability of target detection, low detection efficiency, difficulty in avoiding signal interference and echo distortion, and insufficient target positioning accuracy.

[0006] In a first aspect, this application provides a target detection method based on high-resolution beamforming, comprising:

[0007] After the cylindrical transmitting array transmits the acoustic signal and the cylindrical receiving array receives the acoustic signal, the received acoustic signals of all elements in all layers of the cylindrical receiving array are combined into the original signal vector.

[0008] The beamforming weighted vector of the received acoustic signal is converted into a Kronecker product. Based on the Kronecker product and the joint probability density model, a negative log-likelihood function is constructed. The vectors of azimuth and pitch dimensions are optimized through preset directional constraints to form a high-resolution beam.

[0009] After forming a high-resolution beam, correlation analysis is performed on the echo signals reflected from the target at various azimuth and elevation angles and the transmitted acoustic signals under different time delays.

[0010] Based on the correlation analysis results, a three-dimensional matrix is ​​constructed, which includes the distance, azimuth, and elevation angles of the target and the array elements in three-dimensional space. The three-dimensional matrix is ​​then normalized to eliminate signal strength differences, and the target detection and positioning results are output.

[0011] Optionally, the step of converting the beamforming weighted vector of the received acoustic signal into a Kronecker product, constructing a negative log-likelihood function based on the Kronecker product and the joint probability density model, and optimizing the azimuth and elevation dimension vectors through preset directional constraints to form a high-resolution beam includes:

[0012] Based on the structural characteristics of the concentric rings corresponding to the cylindrical transmitting array, the beamforming weighting vector is decomposed into vectors of azimuth and elevation dimensions, and the beamforming weighting vector is represented by the Kronecker product of the azimuth and elevation dimension vectors.

[0013] The output signal of the beamformer is obtained by combining the conjugate transpose of the beamforming weighting vector with the original signal vector.

[0014] Given that the output signal conforms to a Gaussian distribution, a probability density function corresponding to each time frame is constructed, and the probability density functions corresponding to the time frames are combined to obtain a joint probability density model.

[0015] The joint probability density model is transformed to obtain the negative log-likelihood function, which quantifies the deviation between the output signal and the Gaussian distribution.

[0016] The correspondence between the steering vectors of the cylindrical receiving array in the azimuth dimension and the elevation dimension and the beamforming weight vectors is used to construct a preset direction constraint condition, which indicates that the response of the main lobe of the beam in the target direction is a preset threshold value.

[0017] In combination with the preset direction constraint condition, a Lagrange function is obtained by minimizing a negative log-likelihood function.

[0018] The minimum value of the Lagrange function is iteratively solved by alternately optimizing the vectors in the azimuth dimension and the elevation dimension until the main lobe width and the sidelobe level reach a preset performance threshold value, so as to form a high-resolution beam.

[0019] Optionally, the alternately optimizing the vectors in the azimuth dimension and the elevation dimension comprises:

[0020] Without adjusting the vectors in the elevation dimension, a diagonal loading coefficient is introduced to revise the first covariance matrix to obtain a revised first covariance matrix, and the vectors in the azimuth dimension are optimized by using the revised first covariance matrix.

[0021] Alternatively, without adjusting the vectors in the azimuth dimension, a diagonal loading coefficient is introduced to revise the second covariance matrix to obtain a revised second covariance matrix, and the vectors in the elevation dimension are optimized by using the revised second covariance matrix.

[0022] Optionally, in the process of iteratively solving the minimum value of the Lagrange function until the main lobe width and the sidelobe level reach the preset performance threshold value, the method further comprises:

[0023] The diagonal loading coefficient is dynamically adjusted according to the noise power of the received signal.

[0024] After the diagonal loading coefficient is dynamically adjusted, if the main lobe width and the sidelobe level of the optimized vectors in the azimuth dimension and the elevation dimension do not reach the preset performance threshold value, a genetic algorithm is used to randomly disturb the optimized vectors in the azimuth dimension and the elevation dimension until the preset performance threshold value is reached or the maximum number of iterations is reached.

[0025] Optionally, the correlation analysis of the echo signals reflected by the target at different azimuth angles and elevation angles and the transmitted acoustic wave signals at different time delays comprises:

[0026] The echo signals and the transmitted acoustic wave signals are subjected to wavelet transform to extract time delay features of different frequency bands.

[0027] Based on the time delay features, a sparse time delay spectrum is reconstructed by using a compressive sensing technology to obtain time delay distribution data.

[0028] Based on the time delay distribution data, in combination with real-time sound velocity profile data of an environment where the target is located, a mapping relationship between time delay and distance is dynamically corrected to generate a target distance mapping table;

[0029] Based on the target distance mapping table, a parallel calculation is performed on a traversal process of the azimuth angle and the elevation angle by a GPU cluster, and in the parallel calculation process, each GPU node is assigned to process a mapping relationship between distance and signal strength between corresponding azimuth angle and elevation angle to obtain a correlation analysis result.

[0030] Optionally, based on the time delay feature, a sparse time delay spectrum is reconstructed by a compressive sensing technology to obtain the time delay distribution data, including:

[0031] The time delay feature is decomposed into sparse components at multiple scales, and the sparse components at each scale are independently sparse coded to generate a multi-scale sparse coefficient matrix;

[0032] According to the signal-to-noise ratio of the sparse components at each scale, a reconstruction weight is dynamically assigned, and the multi-scale sparse coefficient matrix is weighted and fused;

[0033] Based on the weighted fusion result, a sparse time delay spectrum is reconstructed by an orthogonal matching pursuit algorithm;

[0034] The reconstructed time delay spectrum is analyzed for energy focusing degree, and if the focusing degree is lower than a preset focusing degree threshold, the scale level of sparse decomposition is adjusted, and the decomposition, coding, assignment, weighted fusion, reconstruction and analysis steps are re-executed until time delay distribution data meeting a preset resolution requirement is generated.

[0035] Optionally, based on the time delay distribution data, in combination with real-time sound velocity profile data of an environment where the target is located, a mapping relationship between time delay and distance is dynamically corrected to generate a target distance mapping table, including:

[0036] According to the environmental temperature data and the depth data, a sound velocity profile function of the environment where the target is located is calculated;

[0037] The sound velocity profile function is solved by a finite element layering method to obtain a correction coefficient for correcting the mapping relationship;

[0038] The mapping relationship between time delay and distance is dynamically corrected by using the correction coefficient to generate the target distance mapping table.

[0039] In a second aspect, the application provides a cylindrical array sonar system, including: a cylindrical transmitting array, a cylindrical receiving array and a controller, the cylindrical transmitting array and the cylindrical receiving array are coaxially aligned;

[0040] The cylindrical transmitting array is composed of multiple concentric rings, each of which uniformly distributes multiple array elements as transmitting units for transmitting acoustic wave signals.

[0041] The cylindrical receiving array is composed of multiple concentric rings, each of which uniformly distributes multiple array elements as receiving units for receiving acoustic wave signals.

[0042] The controller is configured to control the cooperative work of the cylindrical transmitting array and the cylindrical receiving array, and perform the target detection method based on high-resolution beamforming according to any one of the first aspect.

[0043] Optionally, the transmitting units and the receiving units are piezoelectric sensors, the radius of the rings in the cylindrical transmitting array is different from the radius of the rings in the cylindrical receiving array, and the vertical spacing between the rings in the cylindrical transmitting array is different from the vertical spacing between the rings in the cylindrical receiving array, so as to reduce the mutual interference between the transmitting acoustic wave signals and the receiving acoustic wave signals.

[0044] Optionally, the multiple concentric rings in the cylindrical transmitting array include a first main concentric ring and a first backup concentric ring, and the multiple concentric rings in the cylindrical receiving array include a second main concentric ring and a second backup concentric ring.

[0045] In a third aspect, the present application provides a computing device, comprising a processing component and a storage component; the storage component stores one or more computer instructions; the one or more computer instructions are used to be called and executed by the processing component, so as to realize the target detection method based on high-resolution beamforming according to any one of the first aspect.

[0046] In a fourth aspect, the present application provides a computer storage medium, which stores a computer program; when the computer program is executed by a computer, the target detection method based on high-resolution beamforming according to any one of the first aspect is realized.

[0047] The application provides a target detection method based on high-resolution beam forming, which comprises the following steps: after a cylindrical transmitting array transmits a sound wave signal and a cylindrical receiving array receives the sound wave signal, combining the received sound wave signals of all elements of all layers of the cylindrical receiving array into a raw signal vector; converting a beam forming weight vector of the received sound wave signal into a Kronecker product, constructing a negative log-likelihood function based on the Kronecker product and a joint probability density model, and optimizing the azimuth dimension and the elevation dimension vector through a preset direction constraint condition to form a high-resolution beam; after the high-resolution beam is formed, performing correlation analysis on the echo signals reflected by the target at different azimuth angles and elevation angles and the transmitted sound wave signals at different time delays; constructing a three-dimensional matrix containing the distance, azimuth angle and elevation angle of the target in the three-dimensional space and the elements of the matrix, and performing normalization processing on the three-dimensional matrix to eliminate the signal intensity difference, and outputting a target detection positioning result.

[0048] The application has the following beneficial effects: (1) through correlation analysis and matrix normalization processing of the echo signals, high-precision target detection can be realized in a complex underwater environment; (2) through the decomposition technology of the Kronecker product, the computational complexity is greatly reduced, so that the method can efficiently process large-scale array data and adapt to the demand of real-time target detection; (3) through normalization processing, the signal intensity difference caused by different angles and distances is eliminated, and the stability and consistency of the system in different detection environments are enhanced; (4) by combining the distance, azimuth and elevation angle information of the target, the embodiment of the application can provide a complete three-dimensional target detection matrix, which is suitable for underwater sonar, air sonar and other three-dimensional space detection applications.

[0049] Further, the method decomposes the beam forming weight vector into the Kronecker product form of the azimuth and elevation dimensions based on the concentric ring structure of the cylindrical array, combines the conjugate transposed weight vector and the raw signal to generate an output signal; defines that the output signal conforms to the Gaussian distribution, constructs a time frame joint probability density model and converts it into a negative log-likelihood function to quantify the distribution deviation; constructs a Lagrange function by combining the direction constraint condition, iteratively solves the minimum value by alternately optimizing the azimuth and elevation vectors, until the main lobe width and the sidelobe level meet the preset threshold, and forms a high-resolution beam. Through azimuth-elevation dimension decoupling optimization and covariance matrix diagonal loading, the main lobe width is compressed and the sidelobe level is suppressed, and the azimuth and elevation resolutions are improved; by combining the probability density modeling and the direction constraint condition, the robustness to noise and interference is enhanced while the target direction response is stable, and finally the high-precision joint estimation of the target azimuth, elevation and distance parameters is realized, and the reliability and accuracy of the three-dimensional positioning in a complex environment are effectively improved.

[0050] These and other aspects of the present application will become more apparent from the following description. BRIEF DESCRIPTION OF DRAWINGS

[0051] In order to more clearly illustrate the technical solutions in the embodiments of the present application or the prior art, the following will briefly introduce the drawings needed to be used in the description of the embodiments or the prior art. Obviously, the drawings in the following description are some embodiments of the present application, and other drawings can be obtained by those skilled in the art without any creative effort on the basis of these drawings.

[0052] Figure 1 A structural schematic diagram of a cylindrical array sonar system 200 provided by an embodiment of the present application;

[0053] Figure 2 A flowchart of a target detection method 100 based on high-resolution beam forming provided by an embodiment of the present application;

[0054] Figure 3 A structural schematic diagram of a cylindrical receiving array and azimuth distance matrix provided by an embodiment of the present application;

[0055] Figure 4 A flowchart of another target detection method based on high-resolution beam forming provided by an embodiment of the present application;

[0056] Figure 5 A structural schematic diagram of a computing device provided by an embodiment of the present application. DETAILED DESCRIPTION

[0057] In order to make the person skilled in the art better understand the present application, the technical solutions in the embodiments of the present application will be described clearly and completely in the following with reference to the drawings in the embodiments of the present application.

[0058] In some of the descriptions in the specification and claims of the present application and the above-mentioned drawings, a plurality of operations appearing in a specific order are included, but it should be clearly understood that these operations can be executed or in parallel without the order appearing in the text, and the serial numbers of the operations such as 11, 12, etc. are only used to distinguish different operations, and the serial numbers themselves do not represent any execution order. In addition, these processes can include more or fewer operations, and the operations can be executed in sequence or in parallel. It should be noted that the descriptions of "first", "second" and the like in this paper are used to distinguish different messages, devices, modules, etc., and do not represent the sequence, nor limit the "first" and "second" to be different types.

[0059] With reference to the drawings of the embodiments of the present application, the technical solutions in the embodiments of the present application will be clearly and completely described. Obviously, the described embodiments are only a part of the embodiments of the present application, rather than all the embodiments of the present application. Based on the embodiments of the present application, all other embodiments obtained by those skilled in the art without creative work fall within the scope of the present application.

[0060] To solve the problems of insufficient noise suppression, signal distortion and echo interference in the prior art, and the problems of low target detection accuracy and reliability, low detection efficiency, difficult-to-avoid signal interference and echo distortion, and insufficient target positioning accuracy, the embodiments of the present application provide a target detection method based on high-resolution beam forming, which adopts the following concept: first, based on the multi-layer array element structure of the cylindrical receiving array, the sound wave signals received by each layer are integrated into an original signal vector; on this basis, the Kronecker integral resolution characteristics of the beam forming weighting vector are utilized, a negative log-likelihood function is constructed in combination with a joint probability density model, and the weighting vectors in the azimuth / elevation dimensions are alternately optimized through the directional constraint condition, so as to realize high-resolution beam forming with main lobe width compression and side lobe suppression; further, the correlation analysis of the transmitted signal and the target echo under different time delays is combined to extract the target distance information, and a three-dimensional space matrix fusing the azimuth angle, the elevation angle and the distance is constructed; finally, the signal intensity difference is eliminated through normalization processing, and the high-resolution, multi-dimensional joint optimized target three-dimensional positioning result is output.

[0061] Figure 1 A structural schematic diagram of a cylindrical array sonar system 200 provided by the embodiments of the present application is shown in FIG. 2, which includes: Figure 1

[0062] a cylindrical transmitting array 210, a cylindrical receiving array 220 and a controller, the cylindrical transmitting array and the cylindrical receiving array are coaxial 230. The cylindrical transmitting array is composed of a plurality of concentric circular rings 211 with a radius R1, the number of the concentric circular rings 211 with the radius R1 is M1, and each circular ring uniformly distributes a plurality of array elements as transmitting units for transmitting sound wave signals, as shown by 240 in FIG. 2. The cylindrical receiving array is composed of a plurality of concentric circular rings 212 with a radius R2, the number of the concentric circular rings 212 with the radius R2 is M2, and each circular ring uniformly distributes a plurality of array elements as receiving units for receiving sound wave signals, in other words, for receiving beams, as shown by 250 in FIG. 2. The controller is used to control the cooperative work of the cylindrical transmitting array and the cylindrical receiving array, and to execute the above-mentioned target detection method based on high-resolution beam forming. Figure 1 Figure 1

[0063] ​​​In one possible embodiment, both the transmitting unit and the receiving unit are piezoelectric sensors. The radius R1 of the rings in the cylindrical transmitting array is different from the radius R2 of the rings in the cylindrical receiving array; for example, R1 is greater than R2. The vertical spacing h between the rings in the cylindrical transmitting array is different from the vertical spacing h2 between the rings in the cylindrical receiving array to reduce mutual interference between the transmitted and received acoustic signals.

[0064] Specifically, in this embodiment, the cylindrical array sonar system consists of a cylindrical transmitting array and a cylindrical receiving array. More specifically, the cylindrical transmitting array and the cylindrical receiving array are configured with different radii and vertical separation in space, thereby effectively reducing interference and improving target detection accuracy.

[0065] In this embodiment of the application, for the sake of brevity, the cylindrical transmitting array 210 can be simply referred to as the transmitting array, and the cylindrical receiving array 220 can be simply referred to as the receiving array, with a radius of... The concentric rings 211 can be understood as the ring structure of the transmission array, the circular ring array, etc., with a radius of . The concentric ring 212 can be understood as the ring structure of the receiving array, the circular array, etc.

[0066] In this embodiment, in each ring structure of the transmitting array, the ring array of the transmitting array is composed of... It consists of several transmitting units. These transmitting units are uniformly distributed within a radius of [missing information]. A ring array is formed on the annulus, allowing the transmission array to be constructed. Each transmitting unit is an independent transmitter, such as a piezoelectric sensor, used to convert electrical signals into acoustic signals for transmission. In this ring array, all transmitting units are evenly spaced, meaning the distance between array elements is equal, to ensure uniformity and coverage of signal transmission.

[0067] The structure of the transmitting array consists of multiple arrays with the same radius. It consists of a circular array, and the number of circular arrays is Each circular array has a spacing between it and the previous circular array, denoted as . That is, each circular array is parallel to the central axis of the previous circular array, and the vertical spacing is... All the circular arrays are structurally coaxially aligned, meaning that all the circular arrays within the transmitting array share a central axis. The total height of the transmitting array is... That is, the vertical distance between all the circular arrays multiplied by the number of circular arrays. The entire transmitting array consists of... It consists of several transmitting units, providing a large coverage area and stronger signal transmission capability.

[0068] In this embodiment, in each annular structure of the receiving array, the circular annular array of the receiving array is composed of receiving units. These receiving units are uniformly distributed on a circular ring with a radius of , which can form a circular annular array of the receiving array. Each receiving unit is also an independent receiver, such as a piezoelectric sensor, for receiving acoustic signals reflected from the target and converting them into electrical signals. In this circular annular array, the spacing of all receiving units is uniform, i.e., the distance between array elements is equal, to ensure that the received signals have good coverage and uniformity.

[0069] The structure of the receiving array is composed of multiple identical circular annular arrays with a radius of , and the number of circular annular arrays is . There is a spacing between each circular annular array and the previous circular annular array, which is set to . That is, each circular annular array is parallel to the central axis of the previous circular annular array, and the vertical spacing is . All circular annular arrays are coaxially aligned in structure, i.e., all circular annular arrays in the receiving array share a central axis, ensuring consistency in receiving signals. The total height of the receiving array is , i.e., the vertical distance between all circular annular arrays multiplied by the number of circular annular arrays. The entire receiving array is composed of receiving units, which can effectively receive echo signals from different directions and provide a wide range of detection.

[0070] To provide the relative positions of the transmitting array and the receiving array, the embodiments of the present application place both the transmitting array and the receiving array on the same cylindrical structure, and their central axes are coincident, i.e., coaxially aligned. The transmitting array is placed on a larger radius , while the receiving array is placed on a smaller radius , i.e. . This configuration can make the transmitting array cover a wider area, while the receiving array is more concentrated, which helps to improve the accuracy of receiving echo signals. The vertical positions of the transmitting array and the receiving array are different, with the receiving array located below the transmitting array, and the receiving array is located at the bottom of the cylinder, so that the reflected echo signals can be transmitted from the radiation area of the transmitting array to the receiving array. By separating the transmitting array and the receiving array and adjusting their positions, signal interference between the two arrays can be effectively reduced, and the target detection capability of the cylindrical array sonar system can be improved.

[0071] In a possible embodiment, the plurality of concentric circular rings in the cylindrical transmitting array includes a first main concentric circular ring and a first backup concentric circular ring, and the plurality of concentric circular rings in the cylindrical receiving array includes a second main concentric circular ring and a second backup concentric circular ring. To solve the problem of poor signal quality of different layers, the embodiment of the application can perform main-backup switching control.

[0072] Figure 2 A flowchart of a target detection method 100 based on high-resolution beam forming provided by the embodiment of the application is shown in FIG. 1, which is implemented based on the cylindrical array sonar system, and the method includes the following steps: Figure 2

[0073] 110. The system controls all transmitting elements of the cylindrical transmitting array to simultaneously transmit a signal, which is an acoustic wave signal.

[0074] 120. The signal receiving includes the following procedures: step 121, the system waits to avoid self-interference, and step 122, all elements of the cylindrical receiving array receive a signal.

[0075] In this step, the cylindrical receiving array adopts a coaxial configuration, and after the waiting time for avoiding self-interference, all receiving elements synchronously collect echo signals of target reflection and convert them into electrical signals.

[0076] 130. The cylindrical array reduces complexity waveform forming, including the following procedures: step 131, calculating a weight vector , step 132, calculating a weight vector , step 133, calculating a variance , and step 134, scanning an elevation angle and an azimuth angle to obtain a beam forming signal.

[0077] 140. Imaging matrix construction, including the following procedures: step 141, constructing a three-dimensional azimuth range matrix containing distance, azimuth and elevation information, step 142, calculating a normalized azimuth matrix, and step 143, performing target detection. The embodiment of the application eliminates intensity differences through correlation analysis and normalization processing, and finally realizes high-precision target detection. The specific description of the above procedures can refer to another target detection method based on high-resolution beam forming.

[0078] In summary, the low-complexity and high-resolution receiving beam forming is realized by simultaneously transmitting signals by elements in the sonar transmitting array, and the system adopts the low-complexity and high-resolution receiving beam forming method proposed in the embodiment, which is particularly suitable for efficient target detection in the cylindrical array sonar system. The method decomposes the beam forming weight vector, thereby effectively reducing the calculation complexity, while maintaining a high resolution in the signal processing process.

[0079] ​The embodiments of the present application can be based on the above method flow, and the following is described for the low-complexity high-resolution receive beam forming flow and the construction and normalization flow of three-dimensional range-angle matrix respectively.

[0080] (1) The low-complexity high-resolution receive beam forming flow is as follows:

[0081] S1, the embodiments of the present application first perform the representation of the received signal, specifically, the signal of the m-th array element in the n-th time frame is defined as xm,n, and the signals of all array elements can be stacked to form a vector of dimension N, or a matrix of dimension N*N can be used to represent the signals, wherein the element in the m-th row and the n-th column is xm,n.

[0082] S2, the beam forming weight vector is designed, specifically, in order to reduce the calculation complexity, the beam forming weight vector w is defined as the Kronecker product of two vectors: , wherein w1 is a vector of dimension N, and w2 is a vector of dimension M.

[0083] S3, the output of the beam former, specifically: the output signal y of the beam former is y = Xw. , wherein wH represents the conjugate transpose of the weight vector.

[0084] S4, the output signal is defined, specifically, the embodiments of the present application define that the output signal y of the beam former obeys the Gaussian distribution. The probability density function is as follows: , wherein σ2 is the variance to be estimated at the n-th time frame.

[0085] S5, the joint probability density function is constructed, specifically, if T time frames of data are collected for beam forming, the output joint probability density is as follows:

[0086] S6, the negative log-likelihood function is constructed, and an exemplary negative log-likelihood function is as follows:

[0087] ​​​​​​​​​​​​​​​​​​​​​​​S7, determine the constraint condition of target detection, specifically, the direction angle (such as 311 in FIG. 3) Figure 3 , the elevation angle (such as 312 in FIG. 3) , and the pitch angle (such as 313 in FIG. 3) Figure 3 of the target detection relative to the reference (such as 301 in FIG. 3) and the origin (such as 302 in FIG. 3) Figure 3 , require that the weighted vector satisfy: .

[0088] wherein, is the steering vector of the cylindrical array at the direction angle and the pitch angle , is the steering vector of the cylindrical array at the direction angle , and is the steering vector of the cylindrical array at the pitch angle , which can be expressed as: .

[0089] S8, construct a Lagrangian function, specifically, considering the direction constraint condition, minimize the negative log-likelihood function to obtain the Lagrangian function: .

[0090] S9, optimize the weighted vector, specifically, including the following steps S91 and S92, wherein S91, by fixing other values, optimize to obtain the beamformer. Wherein, the formula used for optimizing is:

[0091] ;

[0092] represents the revised covariance matrix, and the calculation formula is:

[0093] ;

[0094] wherein, is an identity matrix, is a diagonal loading coefficient, and by increasing , the stability of the beamformer can be further increased.

[0095] S92, fix other values to optimize to obtain the beamformer, wherein the formula used for optimizing is:

[0096] ;

[0097] wherein, represents the revised covariance matrix, the calculation formula is:​​

[0098] ;

[0099] S10, optimizing scalar , in particular, optimizing the scalar to obtain . .

[0100] S11, at the first calculation S91, the fixed beamformer can be used to initialize , and the revised covariance matrix is calculated with .

[0101] Construction and normalization of three-dimensional range and bearing matrix, the embodiment proposes a target range and bearing matrix construction method based on received beamforming and signal correlation analysis, which is suitable for target detection in cylindrical array sonar system. The method gradually constructs the three-dimensional matrix of the target through the correlation analysis between the echo signal and the transmitted signal, which contains the range, bearing and pitch information of the target, and further improves the detection accuracy and resolution of the system.

[0102] (2) The construction and normalization process of the three-dimensional range and bearing matrix is as follows:

[0103] S21, received beamforming and construction of target range matrix, including S211 and S212.

[0104] Wherein, S211, for each given bearing angle and pitch angle , the received signals of all elements in the array are combined in the above manner to perform received beamforming to enhance the target signal. Through the beamforming process, the echo signal in the specified direction is obtained.

[0105] S212, for each bearing angle and pitch angle , the correlation analysis of echo signal and transmitted signal is performed at different time delays . By calculating the cross-correlation of echo signal and transmitted signal at different time delays , and using the sound velocity to calculate the echo intensity at different distances.

[0106] S22, construction of three-dimensional range and bearing matrix, including S221 and step 222.

[0107] Wherein, S221, for each bearing angle and pitch angle , the range and bearing information of the target is obtained through the correlation analysis between the echo signal and the transmitted signal. Specifically, by calculating the correlation of echo signal and transmitted signal, the signal reflection intensity in different directions can be obtained.

[0108] S222, for each azimuth angle The calculated reflection intensity information and the corresponding distance values are combined to form a distance and bearing matrix of the target (e.g., 321 in FIG. 3). Figure 3 The similar operation is performed for each elevation angle to obtain the reflection intensity of the target at different elevation angles. Finally, the distance and bearing information are combined to obtain a three-dimensional matrix containing the position, bearing angle and elevation angle of the target in three-dimensional space (e.g., 320 in FIG. 3). Figure 3

[0109] S23, normalization of the three-dimensional distance and bearing matrix, including S231 and S232.

[0110] wherein S231, for each elevation angle , the embodiments of the present application calculate the sum of the echo signal intensity (i.e., the corresponding element value in the distance and bearing matrix) at different angles under a given distance value. Specifically, for each distance position, the embodiments of the present application will accumulate the echo intensity at all azimuth angles to obtain the total intensity value at that position. Then, the total intensity value is divided by the number of angles (i.e., the number of azimuth angles of 360 degrees) to obtain the average echo intensity of the distance position at the elevation angle.

[0111] S232, after completing the average value calculation of the echo intensity, the embodiments of the present application divide the original echo intensity matrix by the average value to obtain the normalized matrix. Through this normalization processing, the intensity difference at different angles and distances can be eliminated, so that the values in the matrix are in a unified dimension in each direction. By normalizing the three-dimensional distance and bearing matrix, the embodiments of the present application can obtain a standardized matrix, so that the reflection intensity of the target is more comparable. This normalization step eliminates the intensity difference caused by different angles or distances, thereby improving the accuracy and consistency of target detection.

[0112] In summary, the embodiments of the present application propose a target distance and bearing matrix construction method based on received beam forming and signal correlation analysis, which is suitable for target detection in a cylindrical array sonar system. This method gradually constructs a three-dimensional matrix of the target containing the distance, bearing and elevation information of the target through correlation analysis between the echo signal and the transmitted signal, further improving the detection accuracy and resolution of the system.

[0113] Figure 4 Another flowchart of a target detection method based on high-resolution beam forming provided by the embodiments of the present application is shown in FIG. 4, which includes the following steps: Figure 4

[0114] ​​S41, after the cylindrical transmitting array transmits the sound wave signal and the cylindrical receiving array receives the sound wave signal, the received sound wave signals of all array elements of all layers of the cylindrical receiving array are combined into an original signal vector.

[0115] S42, the beamforming weight vector of the received sound wave signal is converted into a Kronecker product, a negative log-likelihood function is constructed based on the Kronecker product and a joint probability density model, and a vector in the azimuth dimension and the elevation dimension is optimized through a preset direction constraint condition to form a high-resolution beam.

[0116] S43, after the high-resolution beam is formed, the echo signal reflected by the target at each azimuth and elevation is analyzed in correlation with the transmitted sound wave signal at different time delays.

[0117] S44, according to the correlation analysis result, a three-dimensional matrix containing the distance, azimuth and elevation of the target in the three-dimensional space and the array element is constructed, and the three-dimensional matrix is normalized to eliminate the difference in signal intensity, and the target detection and positioning result is output.

[0118] The following is a specific example: after the cylindrical transmitting array synchronously transmits the sound wave signal, the receiving array captures the echo reflected by the target, and arranges all array element signals by layer and position into a 48-dimensional original signal vector . By decomposing the beamforming weight vector into the Kronecker product of the azimuth dimension vector and the elevation dimension vector , and constructing a negative log-likelihood function based on the definition of Gaussian distribution, combining the direction constraint condition, and updating and through Lagrange optimization iteration, a high-resolution beam is generated. Subsequently, the cross-correlation value of the beam signal and the transmitted signal at different time delays τ is calculated by the existing cross-correlation function, the time delay corresponding to the maximum cross-correlation value is selected, for example, the time delay τ = 0.1 s, the sound speed in water is 1500 m / s, and the final actual distance is (1500 x 0.1) / 2 = 75 m. A three-dimensional matrix containing azimuth, elevation and distance is constructed, and the target three-dimensional coordinates are output after the three-dimensional matrix is normalized to eliminate environmental interference.

[0119] By performing S41-S44, the embodiment of the application integrates all array element signals of the receiving array, constructs a high-density original vector, combines Kronecker integral decomposition weighted vector and probability model optimization, reduces the calculation complexity and improves the beam resolution; sub-meter distance accuracy and three-dimensional information fusion are realized through time delay correlation analysis; finally, a normalized three-dimensional matrix is constructed, environmental interference is eliminated, and target three-dimensional coordinates are output, thereby improving the signal-to-noise ratio. The whole process reduces the operation burden while realizing high-precision, strong anti-interference three-dimensional target detection, and is especially suitable for high-resolution imaging and real-time positioning requirements in complex underwater environments.

[0120] In a possible embodiment, S42, the beamforming weighted vector of the received acoustic wave signal is converted into a Kronecker product, a negative log-likelihood function is constructed based on the Kronecker product and a joint probability density model, and the vectors in the azimuth dimension and the elevation dimension are optimized through a preset direction constraint condition to form a high-resolution beam, comprising:

[0121] Step a1, according to the structural characteristics of the concentric circular ring corresponding to the cylindrical transmitting array, the beamforming weighted vector is decomposed into vectors in the azimuth dimension and the elevation dimension, so as to represent the beamforming weighted vector by using the Kronecker product of the vectors in the azimuth dimension and the elevation dimension.

[0122] Step a2, the conjugate transpose of the beamforming weighted vector is combined with the original signal vector to obtain the output signal of the beamformer.

[0123] Step a3, in the case that the output signal conforms to the Gaussian distribution, the probability density function corresponding to each time frame is constructed, and the joint probability density model is obtained by combining the probability density functions corresponding to the time frames.

[0124] Step a4, the joint probability density model is transformed to obtain the negative log-likelihood function, so as to quantify the deviation between the output signal and the Gaussian distribution.

[0125] Step a5, the correspondence between the steering vectors of the cylindrical receiving array in the azimuth dimension and the elevation dimension and the beamforming weighted vector is constructed to form a preset direction constraint condition, and the direction constraint condition indicates that the response of the beam main lobe in the target direction is a preset threshold value.

[0126] Step a6, the negative log-likelihood function is minimized in combination with the preset direction constraint condition to obtain a Lagrange function.

[0127] Step a7, the minimum value of the Lagrange function is iteratively solved by alternately optimizing the vectors in the azimuth dimension and the elevation dimension until the beam main lobe width and the sidelobe level reach the preset performance threshold, so as to form a high-resolution beam.

[0128] The following is a specific example: after the cylindrical transmitting array synchronously transmits the sound wave signal, the receiving array captures the echo reflected by the underwater unmanned submarine, arranges all the array element signals in order of layer and azimuth into a 48-dimensional original signal vector . By decomposing the beamforming weight vector into the Kronecker product of the azimuth dimension vector and the elevation dimension vector , and constructing a joint probability density model based on the Gaussian distribution definition, combining the direction constraint condition (30°,10°)=1, the high-resolution beam is finally generated by iteratively optimizing and .

[0129] By performing steps a1-a7, the embodiment of the application realizes high-resolution beamforming through weight vector decomposition, probability modeling, direction constraint optimization and alternating iteration. In underwater unmanned submarine detection, the system accurately locates the three-dimensional coordinates of the underwater unmanned submarine through three-dimensional matrix construction and normalization, and maintains stable output in a strong noise environment.

[0130] In one possible embodiment, step a7, by alternating optimization of the azimuth dimension and the elevation dimension vector, includes:

[0131] Step b1, without adjusting the vector of the elevation dimension, introduce the diagonal loading coefficient to revise the first covariance matrix, get the revised first covariance matrix, and use the revised first covariance matrix to optimize the vector of the azimuth dimension.

[0132] Wherein, the first covariance matrix refers to the signal statistical characteristic matrix of the azimuth dimension, which is used to describe the correlation of the received signal in the horizontal direction. The fixed weight vector based on the elevation dimension and the received signal data are calculated.

[0133] Step b2, or, without adjusting the vector of the azimuth dimension, introduce the diagonal loading coefficient to revise the second covariance matrix, get the revised second covariance matrix, and use the revised second covariance matrix to optimize the vector of the elevation dimension.

[0134] Wherein, the second covariance matrix refers to the signal statistical characteristic matrix of the elevation dimension, which is used to describe the correlation of the received signal in the vertical direction. The fixed weight vector based on the azimuth dimension and the received signal data are calculated.

[0135] The following is a specific example: after the cylindrical transmitting array, synchronously transmits the sound wave signal, the receiving array captures the echo signal reflected by the underwater unmanned submarine, forms a 48-dimensional original signal vector. By step b1, the elevation dimension vector is fixed , a first covariance matrix of the azimuth dimension is calculated, and a diagonal loading coefficient is introduced , revised as , wherein, represents the azimuth dimension, is the first covariance matrix, is the diagonal loading coefficient, is an identity matrix, the azimuth vector is optimized so that the beam main lobe is accurately aligned with the azimuth angle 30°. Then step b2 is performed, and the optimized , a second covariance matrix of the elevation dimension is calculated, and a diagonal loading coefficient is also added , revised as , wherein, represents the elevation dimension, is the second covariance matrix, is the diagonal loading coefficient, is an identity matrix, the elevation vector is optimized so that the beam main lobe is aligned with the elevation angle 10°.

[0136] By performing steps b1-b2, the embodiments of the present application improve the stability and pointing accuracy of beam forming by alternately optimizing the weight vectors of the azimuth and elevation dimensions in combination with the diagonal loading technology. In a complex underwater environment, this method effectively suppresses the interference of multipath reflection and background noise, making the beam main lobe more concentrated and the sidelobe interference lower, thereby realizing high-resolution three-dimensional target positioning. This scheme is particularly suitable for scenarios that require accurate detection of underwater targets, enhancing the reliability and anti-interference ability of the system.

[0137] In a possible embodiment, in the process of iteratively solving the minimum value of the Lagrangian function until the beam main lobe width and sidelobe level reach the preset performance threshold, the method further comprises:

[0138] Step b11, dynamically adjusting the diagonal loading coefficient according to the noise power of the received signal.

[0139] The noise power of the received signal refers to the energy measure of the non-target component in the received signal, which is usually calculated by the signal power of the non-target period or the noise power outside the signal frequency band. The diagonal loading coefficient is used to adjust the parameter of the stability of the covariance matrix, and by adding a fixed value to the diagonal line of the matrix, numerical calculation problems are avoided.

[0140] Step b12, after dynamically adjusting the diagonal loading coefficient, if the beam main lobe width and sidelobe level of the optimized azimuth and elevation vectors do not reach the preset performance threshold, a genetic algorithm is used to randomly perturb the optimized azimuth and elevation vectors until the preset performance threshold is reached or the maximum number of iterations is reached.

[0141] Wherein, the beam main lobe width refers to the angular range of the beam energy concentration region, reflecting the direction resolution. The side lobe level refers to the energy intensity of the beam non-main lobe direction, reflecting the anti-interference ability. Random disturbance refers to introducing random changes to the vector parameters to jump out of the local optimal solution.

[0142] The following is a specific example: in the underwater unmanned submarine detection task, the receiving array detects high noise power, adjusts p from 0.1 to 0.2, optimizes the beam vector after revising the covariance matrix, but the main lobe width is still 1.3°. Subsequently, 10 groups of disturbance vectors are generated by genetic algorithm, and the optimal solution is screened out after 15 iterations, the main lobe width is compressed to 0.7°, and the side lobe level is reduced to-30dB. Finally, combined with the time delay correlation analysis, the three-dimensional coordinates of the underwater unmanned submarine are output, meeting the high-precision detection requirements.

[0143] By performing steps b11~b12, the embodiments of the present application adaptively cope with different noise environments by dynamically adjusting the diagonal loading coefficient, and improve the stability of the beam forming algorithm; combined with the global search ability of genetic algorithm, further optimize the performance of beam main lobe and side lobe. The two work together to realize high-resolution target detection in complex underwater environment, while avoiding the performance bottleneck caused by local optimal solution.

[0144] In a possible embodiment, S43, the echo signals reflected by the target at each azimuth and elevation angle are correlated with the transmitted sound wave signals at different time delays, including:

[0145] Step c1, wavelet transform the echo signal and the transmitted sound wave signal to extract time delay features of different frequency bands.

[0146] Wherein, the wavelet transform refers to a time-frequency analysis technique, which decomposes the signal into sub-signals of different frequency bands through scaling and translation of basis functions, with time resolution and frequency resolution. The echo signal refers to the sound wave signal reflected by the target, which carries the target distance, azimuth and other information. The transmitted sound wave signal refers to the original sound wave signal actively transmitted by the sonar system, which is usually a linear frequency modulation or pulse signal.

[0147] Step c2, based on the time delay feature, reconstruct the sparse time delay spectrum by compressive sensing technology to obtain the time delay distribution data.

[0148] Wherein, the compressive sensing technology refers to a signal sampling and reconstruction method, which uses the sparsity of the signal to recover the complete signal from a small amount of observation data. The sparse time delay spectrum refers to the sparse representation of the time delay distribution data with a small number of non-zero elements under a certain basis. The time delay distribution data refers to a set of signal energy distribution corresponding to different time delays.

[0149] Step c3, based on the time delay distribution data, combined with the real-time sound speed profile data of the environment where the target is located, dynamically correct the mapping relationship between time delay and distance, and generate a target distance mapping table.

[0150] Wherein, the real-time sound speed profile data refers to the real-time measurement value of the sound speed in the water body with depth. The target distance mapping table refers to the corresponding relationship table between time delay and actual distance of the target, which is dynamically adjusted according to the sound speed.

[0151] Step c4, based on the target distance mapping table, the parallel computing of the azimuth and pitch angle is performed by the GPU cluster, and in the parallel computing process, each GPU node is assigned to process the mapping relationship between the distance and the signal intensity between the corresponding azimuth and pitch angle, and the correlation analysis result is obtained.

[0152] Wherein, the GPU cluster refers to a parallel computing platform composed of multiple GPUs, which is suitable for large-scale data parallel processing. The correlation analysis result refers to the matching degree of the signal in each direction and the target distance mapping table, which is used for positioning the target.

[0153] The following is a specific example: in the underwater unmanned submarine detection task, the wavelet transform of the sonar echo extracts multi-frequency time delay features, and the sparse time delay spectrum is reconstructed by compressive sensing to eliminate multi-path interference peaks; combined with the real-time sound speed profile, the target distance mapping table is generated, and the distance is corrected to 75.2m; using GPU cluster parallel scanning 360x61 directions, the heat map is output within 5 minutes, showing that the underwater unmanned submarine is located at azimuth 30°±0.1°, pitch angle 10°±0.1°, distance 75.2m±0.2m, which meets the centimeter-level detection accuracy requirement.

[0154] By performing steps c1-c4, the present application realizes high-resolution, low-latency target three-dimensional positioning through wavelet multi-frequency time delay extraction, compressive sensing sparse reconstruction, sound speed dynamic correction and GPU parallel computing. In complex underwater environment, the system can accurately map the target distance and direction, improve the detection efficiency, and is suitable for underwater small target detection, unmanned submarine detection, frogman detection and other scenes.

[0155] In one possible embodiment, step c2, based on the time delay feature, the sparse time delay spectrum is reconstructed by compressive sensing technology to obtain the time delay distribution data, including:

[0156] Step c21, decompose the time delay feature into sparse components at multiple scales, independently sparse code the sparse components at each scale, and generate a multi-scale sparse coefficient matrix.

[0157] Wherein, the sparse component refers to a few components in the time delay feature that concentrate energy at a certain scale. Sparse coding refers to representing a signal as a linear combination of a sparse basis, retaining the main components. The multi-scale sparse coefficient matrix refers to a matrix composed of the coding results of sparse components at each scale.

[0158] Step c22, dynamically allocate reconstruction weights according to the signal-to-noise ratio of the sparse components at each scale, and perform weighted fusion on the multi-scale sparse coefficient matrix.

[0159] Wherein, the signal-to-noise ratio refers to the ratio of signal power to noise power, used to measure signal quality. The reconstruction weight refers to the fusion weight allocated to the sparse component at each scale according to the signal-to-noise ratio, and the higher the weight, the higher the reliability of the scale.

[0160] Step c23, reconstruct the sparse time delay spectrum based on the weighted fusion result through the orthogonal matching pursuit algorithm.

[0161] Wherein, the orthogonal matching pursuit refers to an iterative greedy algorithm that reconstructs a sparse signal by gradually selecting the optimal basis function. The sparse time delay spectrum refers to a sparse representation of time delay distribution data, containing only a small number of peaks.

[0162] Step c24, analyze the energy focusing degree of the reconstructed time delay spectrum, if the focusing degree is lower than the preset focusing degree threshold, adjust the scale level of sparse decomposition, and re-execute the decomposition, coding, allocation, weighted fusion, reconstruction and analysis steps until the time delay distribution data that meets the preset resolution requirement is generated.

[0163] Wherein, the energy focusing degree refers to the proportion of the main peak energy to the total energy in the time delay spectrum, which measures the time delay resolution. The preset focusing degree threshold refers to the minimum requirement of the energy focusing degree.

[0164] The following is a specific example: in the underwater unmanned submarine positioning task, the time delay feature is decomposed into 3 scales and sparse coded, and the weights are allocated according to the signal-to-noise ratio for fusion; the time delay spectrum is reconstructed by orthogonal matching pursuit, and the main peak τ=0.1s is found; because the focusing degree is insufficient, the scale level is adjusted to 5 layers, and after re-decomposition, the fusion weight is updated to high frequency 0.7, medium frequency 0.2, and low frequency 0.1, and the final reconstructed time delay spectrum focusing degree reaches 78%, the main peak corresponds to the distance of 75.2m, meeting the detection requirements.

[0165] By executing steps c21-c24, the embodiments of the present application improve the resolution and anti-interference ability of the time delay spectrum through multi-scale decomposition, signal-to-noise ratio weighted fusion, and focusing degree feedback adjustment. In complex underwater environment, the system can accurately separate the main peak of the target, the energy focusing degree is improved, and is suitable for high-precision target detection.

[0166] In one possible embodiment, step c3, based on the time delay distribution data, combines the real-time sound speed profile data of the environment where the target is located to dynamically correct the mapping relationship between time delay and distance, and generates a target distance mapping table, including:

[0167] Step c31, according to the environmental temperature data and the depth data, calculates the sound speed profile function of the environment where the target is located.

[0168] Wherein, the sound speed profile function refers to a function describing the change of sound speed with depth, usually represented as c(z), where z is the depth.

[0169] Step c32, solve the sound speed profile function by finite element layering method to obtain the correction coefficient for correcting the mapping relationship.

[0170] Wherein, the finite element layering method refers to dividing the water body into several horizontal layers according to depth, regarding the sound speed in each layer as a constant, and calculating the correction parameter of sound wave propagation path by layering integration. The correction coefficient refers to an adjustment factor describing the sound speed layering effect on the time delay-distance mapping relationship, used to compensate for the influence of non-uniform sound speed.

[0171] Step c33, dynamically correct the mapping relationship between time delay and distance by using the correction coefficient, and generate a target distance mapping table.

[0172] The following is a specific example: in the three-dimensional positioning task of underwater unmanned submarine, the temperature and salinity data of depth 0-100m are collected by conductivity temperature depth sensor to calculate the sound speed profile function; the water body is divided into 100 layers, and the correction coefficient k=0.0658 s / m is calculated by finite element layering method; step c33 dynamically corrects the time delay τ=0.1s for distance, and generates a target distance mapping table. Finally, combined with GPU parallel computing, the underwater unmanned submarine coordinates are output, which reduces the error compared with the traditional method.

[0173] By performing steps c31-c33, the embodiments of the present application improve the accuracy and reliability of time delay and distance conversion through sound speed profile modeling, layering correction calculation and dynamic mapping table generation. In complex underwater environment, the system can adaptively adjust the influence of sound speed propagation path, eliminate the systematic error caused by traditional uniform sound speed definition, and thus more accurately map the target distance. This scheme is suitable for high-precision target detection and ocean scientific research, effectively coping with the challenge of non-uniform distribution of sound speed in practical application, and improving the stability and environmental adaptability of the detection result.

[0174] Figure 1 The cylindrical array sonar system can perform Figure 2 、 Figure 4The target detection method based on high-resolution beamforming of the embodiment shown has no need to be repeated in terms of implementation principle and technical effects. The specific manner in which each module and unit in the cylindrical array sonar system in the above embodiment performs operations has been described in detail in the embodiment related to the method, and thus will not be described in detail here.

[0175] In one possible design, the controller in the above system can be implemented as a computing device, such as a computer. Figure 5 As shown, the computing device can include a storage component 51 and a processing component 52.

[0176] The storage component 51 stores one or more computer instructions, where the one or more computer instructions are called by the processing component 52 for execution.

[0177] The processing component 52 is configured to: after the cylindrical transmitting array transmits the acoustic wave signal and the cylindrical receiving array receives the acoustic wave signal, combine the received acoustic wave signals of all elements of all layers of the cylindrical receiving array into a raw signal vector; convert a beamforming weighting vector of the received acoustic wave signal into a Kronecker product, construct a negative log-likelihood function based on the Kronecker product and a joint probability density model, and optimize a vector in an azimuth dimension and an elevation dimension through a preset direction constraint condition to form a high-resolution beam; after the high-resolution beam is formed, perform correlation analysis on the echo signals reflected by the target at different azimuth angles and elevation angles and the transmitted acoustic wave signals at different time delays; according to the correlation analysis result, construct a three-dimensional matrix including the distance, azimuth angle and elevation angle of the target in the three-dimensional space and the element, and perform normalization processing on the three-dimensional matrix to eliminate the signal intensity difference, and output the target detection positioning result.

[0178] The processing component 52 can include one or more processors to execute the computer instructions to complete all or part of the steps in the above method. Of course, the processing component can also be one or more application-specific integrated circuits (ASICs), digital signal processors (DSPs), digital signal processing devices (DSPDs), programmable logic devices (PLDs), field programmable gate arrays (FPGAs), controllers, microcontrollers, microprocessors or other electronic elements for executing the above method.

[0179] Storage component 51 is configured to store various types of data to support operations at the terminal. The storage component can be implemented from any type of volatile or non-volatile storage device or a combination thereof, such as Random Access Memory (RAM), Static Random-Access Memory (SRAM), Electrically Erasable Programmable Read Only Memory (EEPROM), Erasable Programmable Read Only Memory (EPROM), Programmable Read Only Memory (PROM), Read Only Memory (ROM), magnetic storage, flash memory, magnetic disk, or optical disk.

[0180] Of course, computing devices may also include other components, such as input / output interfaces, display components, communication components, etc.

[0181] Input / output interfaces provide interfaces between processing components and peripheral interface modules, which can be output devices, input devices, etc.

[0182] The communication components are configured to facilitate wired or wireless communication between computing devices and other devices.

[0183] The computing device can be a physical device or an elastic computing host provided by a cloud computing platform. In this case, the computing device can refer to a cloud server, and the aforementioned processing components, storage components, etc., can be basic server resources rented or purchased from the cloud computing platform.

[0184] This application also provides a computer storage medium storing a computer program, which, when executed by a computer, can perform the above-described functions. Figure 2 or Figure 4 The embodiment shown is a target detection method based on high-resolution beamforming.

[0185] Those skilled in the art will clearly understand that, for the sake of convenience and brevity, the specific working processes of the systems, devices, and units described above can be referred to the corresponding processes in the foregoing method embodiments, and will not be repeated here.

[0186] The device embodiments described above are merely illustrative, wherein the units described as separate components can or can not be physically separate, and the components displayed as units can or can not be physical units, i.e., can be located in one place, or can be distributed to multiple network units. Part or all of the modules can be selected to achieve the purposes of the embodiments according to actual needs. Those skilled in the art can understand and implement without creative labor.

[0187] Through the description of the above embodiments, those skilled in the art can clearly understand that the embodiments can be realized by means of software and the necessary general hardware platform, and of course can also be realized by hardware. Based on such understanding, the above technical solutions can be embodied in the form of a software product, which can be stored in a computer readable storage medium, such as a ROM / RAM, a magnetic disk, an optical disk, etc., and includes a number of instructions to make a computer device (which can be a personal computer, a server, or a network device, etc.) execute the methods described in each embodiment or some parts of the embodiments.

[0188] Finally, it should be noted that: the above embodiments are only used to illustrate the technical solutions of the present application, and not to limit them; although the present application has been described in detail with reference to the foregoing embodiments, those skilled in the art should understand that: it can still modify the technical solutions recorded in the foregoing embodiments, or make equivalent replacement for part of the technical features; and these modifications or replacements do not make the essence of the corresponding technical solutions deviate from the spirit and scope of the technical solutions of the embodiments of the present application.

Claims

1. A target detection method based on high resolution beamforming, characterized in that, The method comprises the following steps: After the cylindrical transmitting array transmits the acoustic wave signal and the cylindrical receiving array receives the acoustic wave signal, the received acoustic wave signals of all elements of all layers of the cylindrical receiving array are combined into an original signal vector; The beamforming weight vector of the received acoustic wave signal is converted into a Kronecker product, a negative log-likelihood function is constructed based on the Kronecker product and a joint probability density model, and the vectors in the azimuth dimension and the elevation dimension are optimized through a preset directional constraint condition to form a high-resolution beam; After the high-resolution beam is formed, the correlation analysis is performed on the echo signals reflected by the target at different azimuth angles and elevation angles and the transmitted acoustic wave signals at different time delays; According to the correlation analysis result, a three-dimensional matrix containing the distance, azimuth angle and elevation angle of the target in the three-dimensional space and the element is constructed, and the three-dimensional matrix is normalized to eliminate the signal intensity difference, and the target detection and positioning result is output; The beamforming weight vector of the received acoustic wave signal is converted into a Kronecker product, a negative log-likelihood function is constructed based on the Kronecker product and a joint probability density model, and the vectors in the azimuth dimension and the elevation dimension are optimized through a preset directional constraint condition to form a high-resolution beam, comprising: According to the structural characteristics of the concentric circular rings corresponding to the cylindrical transmitting array, the beamforming weight vector is decomposed into vectors in the azimuth dimension and the elevation dimension, so as to represent the beamforming weight vector by using the Kronecker product of the vectors in the azimuth dimension and the elevation dimension; The conjugate transpose of the beamforming weight vector is combined with the original signal vector to obtain the output signal of the beamformer; In the case that the output signal conforms to the Gaussian distribution, the probability density functions corresponding to the time frames are constructed, and the joint probability density model is obtained by combining the probability density functions corresponding to the time frames; The joint probability density model is transformed to obtain a negative log-likelihood function to quantify the deviation between the output signal and the Gaussian distribution; The corresponding relationship between the steering vectors of the cylindrical receiving array in the azimuth dimension and the elevation dimension and the beamforming weight vector is used to construct a preset directional constraint condition, and the directional constraint condition represents that the response of the main lobe of the beam in the target direction is a preset threshold value; The negative log-likelihood function is minimized in combination with the preset directional constraint condition to obtain a Lagrange function; The minimum value of the Lagrange function is iteratively solved by alternately optimizing the vectors in the azimuth dimension and the elevation dimension until the main lobe width and the sidelobe level reach the preset performance threshold to form a high-resolution beam.

2. The method of claim 1, wherein, The vectors in the azimuth dimension and the elevation dimension are alternately optimized, comprising: Without adjusting the vector in the elevation dimension, a diagonal loading coefficient is introduced to revise the first covariance matrix to obtain a revised first covariance matrix, and the vector in the azimuth dimension is optimized by using the revised first covariance matrix; Or, without adjusting the vector in the azimuth dimension, a diagonal loading coefficient is introduced to revise the second covariance matrix to obtain a revised second covariance matrix, and the vector in the elevation dimension is optimized by using the revised second covariance matrix.

3. The method of claim 2, wherein, In the process of iteratively solving the minimum value of the Lagrange function until the beam main lobe width and side lobe level reach the preset performance threshold, the method further comprises: According to the noise power of the received signal, the diagonal loading coefficient is dynamically adjusted; After dynamically adjusting the diagonal loading coefficient, if the beam main lobe width and side lobe level of the optimized azimuth dimension and elevation dimension vectors do not reach the preset performance threshold, a genetic algorithm is used to randomly perturb the optimized azimuth dimension and elevation dimension vectors until the preset performance threshold is reached or the maximum number of iterations is reached.

4. The method of claim 1, wherein, The correlation analysis of the echo signal reflected by the target at each azimuth and elevation and the transmitted acoustic wave signal at different time delays includes: Wavelet transform is performed on the echo signal and the transmitted acoustic wave signal to extract time delay features of different frequency bands; Based on the time delay features, a sparse time delay spectrum is reconstructed by a compressive sensing technology to obtain time delay distribution data; Based on the time delay distribution data, the mapping relationship between time delay and distance is dynamically corrected in combination with real-time sound speed profile data of the environment where the target is located to generate a target distance mapping table; Based on the target distance mapping table, parallel computing is performed on the traversal process of the azimuth and elevation angles by a GPU cluster, and in the parallel computing process, each GPU node is assigned to process the mapping relationship between the distance and the signal strength between the corresponding azimuth and elevation angles to obtain the correlation analysis result.

5. The method of claim 4, wherein, The correlation analysis of the echo signal reflected by the target at each azimuth and elevation and the transmitted acoustic wave signal at different time delays includes: The time delay features are decomposed into sparse components at multiple scales, and the sparse components at each scale are independently sparse coded to generate a multi-scale sparse coefficient matrix; According to the signal-to-noise ratio of the sparse components at each scale, the reconstruction weight is dynamically allocated, and the multi-scale sparse coefficient matrix is weighted and fused; Based on the weighted fusion result, the sparse time delay spectrum is reconstructed by an orthogonal matching pursuit algorithm; If the focusing degree is lower than the preset focusing degree threshold, the scale level of sparse decomposition is adjusted, and the decomposition, coding, allocation, weighted fusion, reconstruction and analysis steps are re-executed until the time delay distribution data meeting the preset resolution requirement is generated.

6. The method of claim 4, wherein, The correlation analysis of the echo signal reflected by the target at each azimuth and elevation and the transmitted acoustic wave signal at different time delays includes: According to the environmental temperature data and depth data, the sound speed profile function of the environment where the target is located is calculated; The sound speed profile function is solved by a finite element layering method to obtain a correction coefficient for correcting the mapping relationship; The mapping relationship between time delay and distance is dynamically corrected using the correction coefficient to generate a target distance mapping table.

7. A cylindrical array sonar system characterized by, It comprises: A cylindrical transmitting array, a cylindrical receiving array and a controller, the cylindrical transmitting array and the cylindrical receiving array are coaxially aligned; The cylindrical transmitting array is composed of a plurality of concentric circular rings, each circular ring uniformly distributes a plurality of elements as transmitting units for transmitting acoustic wave signals. The cylindrical receiving array is composed of multiple concentric rings, each of which uniformly distributes multiple array elements as receiving units for receiving acoustic wave signals; The controller is configured to control the cooperative work of the cylindrical transmitting array and the cylindrical receiving array, and perform the target detection method based on high-resolution beam forming according to any one of claims 1 to 6.

8. The system of claim 7, wherein, The transmitting unit and the receiving unit are both piezoelectric sensors, the radius of the ring in the cylindrical transmitting array is different from the radius of the ring in the cylindrical receiving array, and the vertical spacing between the rings in the cylindrical transmitting array is different from the vertical spacing between the rings in the cylindrical receiving array, so as to reduce the mutual interference between the transmitting acoustic wave signals and the receiving acoustic wave signals.

9. The system of claim 7 or 8, wherein, The multiple concentric rings in the cylindrical transmitting array include a first main concentric ring and a first backup concentric ring, and the multiple concentric rings in the cylindrical receiving array include a second main concentric ring and a second backup concentric ring.

Images