Iterative fast microphone array design method and system

Abstract

The invention discloses an iterative rapid microphone array design method and system, and the method comprises the following steps: setting microphone array parameters and an arrangement mode of initial basic units, and forming the basic units by employing equilateral triangle sub-arrays; gradually adding equilateral triangle sub-arrays into the whole array in an iteration mode; calculating the output of a point spread function of the array relative to an array focusing direction, calculating a main lobe width and a maximum side lobe level, constructing an optimized objective function of a genetic algorithm, and searching and adding an optimal arrangement position of a group of equilateral triangle sub-arrays based on the genetic algorithm by taking the maximized optimized objective function as a design direction; and judging whether a preset iteration stop condition is met or not, if so, completing the microphone array design, otherwise, continuing to add the equilateral triangle sub-arrays until the preset iteration stop condition is met, and outputting a final microphone array design result. According to the method, rapid microphone array optimization design can be carried out, and the number of array elements in the array can be autonomously controlled.

Description

Technical Field

[0001] The present invention relates to the technical field of sound source imaging, and in particular to an iterative fast microphone array design method and system. Background Art

[0002] Acoustic source imaging technology can aid in fault diagnosis, repair, or structural optimization by locating noise sources in equipment. For example, it can identify discharge areas in transformer boxes for repair or locate vibration zones on aircraft wings to enhance structural stability. However, the performance of acoustic source imaging technology is highly dependent on the configuration and layout of the microphone array. Typical arrays typically have a fixed structure and a limited frequency response range. In complex or highly variable applications, a single microphone array often struggles to simultaneously meet the dual requirements of high-resolution imaging and wideband adaptability. This is especially true when the target frequency range exceeds the array's optimal operating bandwidth, resulting in imaging artifacts or insufficient spatial resolution, leading to inaccurate sound source localization. To address this challenge, the array structure must be redesigned or optimized, a time-consuming and labor-intensive process that also carries high hardware and system costs, especially when a large number of microphone elements and simultaneous acquisition channels are required. Therefore, for complex applications that cannot be met with a single array structure, a new acoustic source imaging array design technology is urgently needed that can rapidly optimize the microphone array structure, adapt to a specific frequency range, and minimize the number of elements while maintaining imaging performance. Summary of the Invention

[0003] In order to overcome the defects and shortcomings of the existing technology, the present invention provides an iterative fast microphone array design method and system. The present invention uses an equilateral triangle subarray as the basic unit, adds a group of basic units based on the results of the previous iteration, takes the main lobe width and maximum sidelobe level in the point spread function of the array as the optimization objective function, determines the equilateral triangle subarray arrangement that can optimize the objective function through a genetic algorithm, and gradually adds basic units to the array in multiple iterations, thereby gradually increasing and controlling the number of array elements, and achieving performance optimization of the array for the target frequency range.

[0004] In order to achieve the above object, the present invention adopts the following technical solutions:

[0005] The present invention provides an iterative fast microphone array design method, comprising the following steps:

[0006] Setting microphone array parameters and the arrangement of the initial basic unit, using an equilateral triangle subarray to form the basic unit, where the three array elements of the equilateral triangle subarray are located at the three vertices of the same equilateral triangle;

[0007] Iteratively adding equilateral triangle sub-arrays to the overall array step by step;

[0008] The array's point spread function output with respect to the array's focusing direction is calculated, along with the main lobe width and maximum sidelobe level. Based on these values, a genetic algorithm optimization objective function is constructed. With maximizing the optimization objective function as the design direction, the genetic algorithm is used to search for the optimal placement of a set of equilateral triangle subarrays.

[0009] Determine whether the preset iteration stop condition is met. If so, complete the microphone array design. Otherwise, continue to add equilateral triangle subarrays until the preset iteration stop condition is met, and output the final microphone array design result.

[0010] As a preferred technical solution, microphone array parameters include: array aperture, limit on the number of array elements in the array, range of array element polar diameter values, range of array element polar angle values, target frequency of array optimization design and expected array performance indicators.

[0011] As a preferred technical solution, the output of the point spread function of the array with respect to the array focusing direction is calculated, which is specifically expressed as:

[0012]

[0013] Among them, k represents the unit direction vector from the geometric center point of the microphone array plane to the scanning point, that is, the unit direction vector of the array focus, k0 represents the direction vector of the unit intensity far-field plane incoming wave, x n represents the position coordinate vector of the nth element in the microphone array, ω represents the sound signal frequency, c represents the sound propagation speed, N represents the number of elements in the array, and j represents an imaginary number.

[0014] As a preferred technical solution, the unit direction vector k of the array focusing direction is expressed as a direction angle and a pitch angle, and the range of the point spread function for extracting the main lobe width and the maximum side lobe level is expressed as:

[0015]

[0016] Where θ represents the direction angle, Indicates the pitch angle.

[0017] As a preferred technical solution, the optimization objective function of the genetic algorithm is constructed according to the main lobe width and the maximum side lobe level, which is specifically expressed as:

[0018] V=MSL-MLW

[0019] Wherein, V represents the optimization objective function of the genetic algorithm, MSL represents the maximum sidelobe level, and MLW represents the mainlobe width.

[0020] As a preferred technical solution, with the maximization of the optimization objective function as the design direction, a genetic algorithm is used to search for the optimal layout position of adding a group of equilateral triangle sub-arrays, specifically including:

[0021] A genetic algorithm is used to randomly select values ​​and find the result that maximizes the optimization objective function from multiple groups of value results. The polar diameter and polar angle of one array element in the equilateral triangle subarray in the optimal layout are returned. The optimization variables of the genetic algorithm are the polar diameter and polar angle of an array element in the equilateral triangle subarray. Based on the geometric property that the geometric center of the equilateral triangle subarray coincides with the array center, the coordinates of the other two array elements are determined by the polar diameter and polar angle of one array element.

[0022] The present invention also provides an iterative fast microphone array design system for implementing the above-mentioned iterative fast microphone array design method, comprising: an initial state setting module, a subarray iterative addition module, an optimization objective function construction module, an iterative stop condition judgment module, and a microphone array design result output module;

[0023] The initial state setting module is used to set the microphone array parameters and the arrangement of the initial basic unit, using an equilateral triangle subarray to form the basic unit, and the three array elements of the equilateral triangle subarray are respectively located at the three vertices of the same equilateral triangle;

[0024] The subarray iterative adding module is used to add equilateral triangle subarrays to the overall array step by step in an iterative manner;

[0025] The optimization objective function construction module is used to construct the optimization objective function of the genetic algorithm, specifically including:

[0026] The array's point spread function output with respect to the array's focusing direction is calculated, along with the main lobe width and maximum sidelobe level. Based on these values, a genetic algorithm optimization objective function is constructed. With maximizing the optimization objective function as the design direction, the genetic algorithm is used to search for the optimal placement of a set of equilateral triangle subarrays.

[0027] The iteration stop condition judgment module is used to judge whether a preset iteration stop condition is met. If the preset iteration stop condition is met, the microphone array design is completed; otherwise, the equilateral triangle subarray is continued to be added until the preset iteration stop condition is met;

[0028] The microphone array design result output module is used to output the final microphone array design result.

[0029] Compared with the prior art, the present invention has the following advantages and beneficial effects:

[0030] The present invention utilizes the geometric properties of equilateral triangle subarrays and array point spread functions during the genetic algorithm calculation process to effectively reduce the amount of calculation, thereby improving the calculation speed and allowing rapid design. It can quickly optimize the microphone array design for the frequency of interest, improve the microphone array's ability to resolve target frequency signals, and can autonomously control the number of array elements in the array, allowing the subtraction of unnecessary array elements to save costs. BRIEF DESCRIPTION OF THE DRAWINGS

[0031] Figure 1 Schematic diagram of the flow of the iterative fast microphone array design method of the present invention;

[0032] Figure 2 A two-dimensional schematic diagram of the main lobe width and maximum side lobe level extracted from the array point spread function of the present invention;

[0033] Figure 3 Schematic diagram of the iterative process of the iterative fast microphone array design method of the present invention. DETAILED DESCRIPTION

[0034] In order to make the purpose, technical solutions and advantages of the present invention more clearly understood, the present invention is further described in detail below with reference to the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are only used to explain the present invention and are not intended to limit the present invention. The step numbers in the following embodiments are only provided for the convenience of explanation and do not limit the order of the steps. The execution order of each step in the embodiments can be adaptively adjusted according to the understanding of those skilled in the art.

[0035] Example 1

[0036] like Figure 1 As shown, this embodiment provides an iterative fast microphone array design method, including the following steps:

[0037] S1: Set the microphone array parameters and the initial basic unit layout;

[0038] In this embodiment, microphone array parameters are first set, including array aperture, number limit of array elements in the array, array element polar diameter, array element polar angle, target frequency of array optimization design, and expected array performance indicators. Among them, the array aperture determines the overall size of the array. For example, when the aperture is limited to no more than 1 meter, the polar diameter of the array element in the equilateral triangle subarray does not exceed 0.5 meters; the number limit of array elements determines the maximum number of array elements allowed in the array, which is used to avoid adding too many array elements during the process of optimizing array performance, resulting in a surge in costs; the target frequency of the array optimization design determines the performance of the optimized array for signals of which frequency; the expected array performance indicators are expressed in MLW and MSL. MLW measures the spatial resolution of the array to the signal. The smaller the MLW, the higher the resolution. MSL measures the degree of image contamination caused by the side lobes formed by the sound source imaging algorithm. The larger the MSL, the lower the degree of contamination.

[0039] In this embodiment, an equilateral triangle subarray is used to form a basic unit. The equilateral triangle subarray as a basic unit includes three array elements, and the three array elements are respectively located at the three vertices of the same equilateral triangle. The equilateral triangle is located in the overall array plane, and its geometric center coincides with the geometric center of the array. That is, the equilateral triangle subarray has rotational symmetry around the array center.

[0040] In this embodiment, with the array center as the origin of the polar coordinate system, the specific arrangement of the equilateral triangle sub-array can be determined by the polar diameter and polar angle of one of the array elements.

[0041] S2: Iteratively add equilateral triangle subarrays to the overall array to improve array design speed and control the number of array elements;

[0042] In this embodiment, an initial basic unit is randomly added based on a genetic algorithm. For example, when designing an array with an aperture not exceeding 1 meter, an equilateral triangle subarray with an element having a polar diameter of 0.25 meters and a polar angle of 0° can be first added. Then, an iterative design is performed. Based on the overall array obtained in the previous iteration, and with maximizing the optimization objective function as the design direction, the genetic algorithm is used to search for the optimal layout position for adding a group of equilateral triangle subarrays. Specifically, the optimization variables of the genetic algorithm are the polar diameter and polar angle of an element in the equilateral triangle subarray. The value ranges of the polar diameter and polar angle are included in the array parameter settings. The genetic algorithm randomly selects values ​​and finds the result that maximizes the optimization objective function from multiple sets of value results, thereby continuously approaching the optimal layout position. The genetic algorithm ultimately returns the polar diameter and polar angle of one element in the equilateral triangle subarray in the optimal layout solution. Utilizing the geometric property that the geometric center of the equilateral triangle subarray at this time coincides with the array center, that is, the coordinate origin coincides, and the three vertices of the triangle have an obvious geometric relationship, the polar diameter and polar angle of one element can be used to determine the coordinates of the other two elements.

[0043] S3: Calculate the output of the array's point spread function with respect to the array's focusing direction, calculate the main lobe width and the maximum side lobe level, and use the main lobe width and the maximum side lobe level in the array's point spread function to form the genetic algorithm optimization objective function, specifically including:

[0044] First calculate the output of the point spread function with respect to the array focusing direction. The specific calculation formula is expressed as:

[0045]

[0046] Where k represents the unit direction vector from the geometric center point O of the microphone array plane to the scanning point X, that is, the unit direction vector of the array focus, k0 represents the direction vector of the unit intensity far-field plane incoming wave, x n represents the position coordinate vector of the nth element in the microphone array, ω represents the sound signal frequency, c represents the sound propagation speed, N represents the number of elements in the array, and j represents an imaginary number. From the calculation formula, we can see that Y is only related to (k-k0) and has nothing to do with the absolute value of k0. Therefore, the "shape" of Y is not affected by the direction of the incoming wave. This feature can be used in subsequent steps to reduce the amount of calculation of the point spread function.

[0047] like Figure 2 As shown, after calculating the output of the point spread function with respect to the array focusing direction, the output of the point spread function with respect to the array focusing direction is converted into decibel expression, the level of the highest peak is normalized to 0 dB, and the width of the area surrounding the -3 dB point on the highest peak is called the main lobe width (MLW); the secondary peak is found, and the difference between the secondary lobe vertex and the main lobe vertex is called the maximum sidelobe level (MSL). In this implementation, MLW and MSL constitute the optimization objective function of the genetic algorithm, which is specifically expressed as:

[0048] V=MSL-MLW

[0049] Where V represents the optimization objective function of the genetic algorithm. The optimization direction is to maximize the objective function of the genetic algorithm and explore the optimal arrangement of the equilateral triangle subarray.

[0050] S4: Determine whether the array performance meets the standard or whether the number of array elements reaches the set upper limit. If the array performance meets the standard or the number of array elements reaches the set upper limit, the array design is completed. Otherwise, return to step S2 to continue adding equilateral triangle subarrays until the array performance meets the standard or the number of array elements reaches the set upper limit. This embodiment iteratively adds basic units to the overall array to improve the array design speed and control the number of array elements. At the same time, the geometric characteristics of the equilateral triangle subarray and the array point spread function are used to reduce the amount of calculation.

[0051] like Figure 3 As shown, multiple iterations are performed until the performance parameters of the overall array meet the standards or the number of array elements in the array reaches an upper limit, thereby avoiding the huge amount of calculation caused by calculating too many basic units in a single time, improving the array design speed and allowing the number of array elements to be controlled;

[0052] In this embodiment, the specific process of reducing computational complexity by utilizing the geometric characteristics of equilateral triangle subarrays and array point spread functions is to first adopt equilateral triangle subarrays as the basic unit. This allows a single genetic algorithm optimization calculation to determine the positions of three array elements, and only requires the coordinate parameters of one array element as input to the genetic algorithm, thereby significantly reducing the complexity of the genetic algorithm calculation and improving the computational speed. Among various central rotationally symmetric subarrays, the greater the number of array elements in the subarray, the more array elements are added in a single iteration, and the lower the control accuracy of the array element number. The smaller the number of array elements in the subarray, the higher the control accuracy of the array element number, but the slower the array element growth rate, requiring more iterations and longer computation time to meet the required array element number. Therefore, after balancing the control accuracy of the array element number and computation time, the equilateral triangle subarray is selected as the basic unit.

[0053] Secondly, considering that the shape of the array point spread function output about the array focusing direction is not affected by the incoming wave direction, the calculation can be simplified to consider the case where the far-field plane incoming wave is perpendicular to the array plane, that is, k0 = (0, 0, -1). At this time, combining the formula for Y with the rotational symmetry of the equilateral triangle subarray, that is, it exhibits a 120° rotation periodicity around the center of the array, for a fixed sound signal frequency ω, it can be obtained:

[0054]

[0055] That is, Y also presents a 120° periodicity around the Z axis that passes vertically through the center of the array. The unit direction vector k in the focusing direction of the array is expressed by the direction angle θ and the pitch angle Therefore, the range of Y needed to extract MLW and MSL is:

[0056]

[0057] can be simplified to:

[0058]

[0059] As can be seen, the output Y of the point spread function is rotationally symmetric about the Z axis and exhibits a 120° periodicity. Therefore, the MSL and MLW indicators can be extracted by simply calculating the point spread function within the azimuth angle range of 0 to 120°. This reduces the original Y calculation workload to one-third, significantly improving the speed of the genetic algorithm calculation and accelerating the overall design process.

[0060] In this embodiment, the geometric characteristics of the equilateral triangle subarray and the array point spread function are utilized to reduce the amount of calculation. The rotational symmetry of the equilateral triangle subarray around the array center is utilized to compress the parameters to be optimized to the coordinates of only a single array element. The periodicity of the array point spread function in the azimuthal dimension is utilized to reduce the point spread function calculation range required for extracting MLW and MSL.

[0061] Example 2

[0062] This embodiment provides an iterative fast microphone array design system for implementing the iterative fast microphone array design method of embodiment 1, the system comprising: an initial state setting module, a subarray iterative addition module, an optimization objective function construction module, an iterative stop condition judgment module, and a microphone array design result output module;

[0063] In this embodiment, the initial state setting module is used to set the microphone array parameters and the arrangement of the initial basic unit. The basic unit is composed of an equilateral triangle subarray, and the three array elements of the equilateral triangle subarray are respectively located at the three vertices of the same equilateral triangle.

[0064] In this embodiment, the subarray iterative adding module is used to gradually add equilateral triangle subarrays to the overall array in an iterative manner;

[0065] In this embodiment, the optimization objective function construction module is used to construct the optimization objective function of the genetic algorithm, specifically including:

[0066] The array's point spread function output with respect to the array's focusing direction is calculated, along with the main lobe width and maximum sidelobe level. Based on these values, a genetic algorithm optimization objective function is constructed. With maximizing the optimization objective function as the design direction, the genetic algorithm is used to search for the optimal placement of a set of equilateral triangle subarrays.

[0067] In this embodiment, the iteration stop condition judgment module is used to judge whether a preset iteration stop condition is met. If the preset iteration stop condition is met, the microphone array design is completed; otherwise, the equilateral triangle subarray is continued to be added until the preset iteration stop condition is met.

[0068] In this embodiment, the microphone array design result output module is used to output the final microphone array design result.

[0069] The above embodiments are preferred implementation modes of the present invention, but the implementation modes of the present invention are not limited to the above embodiments. Any other changes, modifications, substitutions, combinations, and simplifications that do not deviate from the spirit and principles of the present invention should be considered as equivalent replacement methods and are included in the scope of protection of the present invention.

Claims

1. An iterative fast microphone array design method, characterized in that: The steps include: Setting microphone array parameters and the arrangement of the initial basic unit, using an equilateral triangle subarray to form the basic unit, where the three array elements of the equilateral triangle subarray are located at the three vertices of the same equilateral triangle; Iteratively adding equilateral triangle sub-arrays to the overall array step by step; The array's point spread function output with respect to the array's focusing direction is calculated, along with the main lobe width and maximum sidelobe level. Based on these values, a genetic algorithm optimization objective function is constructed. With maximizing the optimization objective function as the design direction, the genetic algorithm is used to search for the optimal placement of a set of equilateral triangle subarrays. Determine whether the preset iteration stop condition is met. If so, complete the microphone array design. Otherwise, continue to add equilateral triangle subarrays until the preset iteration stop condition is met, and output the final microphone array design result.

2. The iterative fast microphone array design method according to claim 1, characterized in that Microphone array parameters include: array aperture, limit on the number of array elements in the array, range of array element polar diameters, range of array element polar angles, target frequency for array optimization design, and expected array performance indicators.

3. The iterative fast microphone array design method according to claim 1, characterized in that Calculate the output of the array's point spread function with respect to the array's focusing direction, specifically expressed as: Among them, k represents the unit direction vector from the geometric center point of the microphone array plane to the scanning point, that is, the unit direction vector of the array focus, k0 represents the direction vector of the unit intensity far-field plane incoming wave, x n represents the position coordinate vector of the nth element in the microphone array, ω represents the sound signal frequency, c represents the sound propagation speed, N represents the number of elements in the array, and h represents an imaginary number.

4. The iterative fast microphone array design method according to claim 3, characterized in that: The unit direction vector k of the array focusing direction is expressed as the direction angle and the elevation angle. The range of the point spread function for extracting the main lobe width and the maximum side lobe level is expressed as: Where θ represents the direction angle, Indicates the pitch angle.

5. The iterative fast microphone array design method according to claim 1, wherein: The optimization objective function of the genetic algorithm is constructed based on the main lobe width and the maximum side lobe level, which is specifically expressed as: V=MSL-MLW Wherein, V represents the optimization objective function of the genetic algorithm, MSL represents the maximum sidelobe level, and MLW represents the mainlobe width.

6. The iterative fast microphone array design method according to claim 1, wherein: With the maximization of the optimization objective function as the design direction, a genetic algorithm is used to search for the optimal layout of a set of equilateral triangle sub-arrays, specifically including: A genetic algorithm is used to randomly select values ​​and find the result that maximizes the optimization objective function from multiple groups of value results. The polar diameter and polar angle of one array element in the equilateral triangle subarray in the optimal layout are returned. The optimization variables of the genetic algorithm are the polar diameter and polar angle of an array element in the equilateral triangle subarray. Based on the geometric property that the geometric center of the equilateral triangle subarray coincides with the array center, the coordinates of the other two array elements are determined by the polar diameter and polar angle of one array element.

7. An iterative fast microphone array design system, characterized in that: The method is used to implement the iterative fast microphone array design method according to any one of claims 1 to 6, comprising: an initial state setting module, a subarray iterative adding module, an optimization objective function construction module, an iterative stop condition judgment module, and a microphone array design result output module; The initial state setting module is used to set the microphone array parameters and the arrangement of the initial basic unit, using an equilateral triangle subarray to form the basic unit, and the three array elements of the equilateral triangle subarray are respectively located at the three vertices of the same equilateral triangle; The subarray iterative adding module is used to add equilateral triangle subarrays to the overall array step by step in an iterative manner; The optimization objective function construction module is used to construct the optimization objective function of the genetic algorithm, specifically including: The array's point spread function output with respect to the array's focusing direction is calculated, along with the main lobe width and maximum sidelobe level. Based on these values, a genetic algorithm optimization objective function is constructed. With maximizing the optimization objective function as the design direction, the genetic algorithm is used to search for the optimal placement of a set of equilateral triangle subarrays. The iteration stop condition judgment module is used to judge whether a preset iteration stop condition is met. If the preset iteration stop condition is met, the microphone array design is completed; otherwise, the equilateral triangle subarray is continued to be added until the preset iteration stop condition is met; The microphone array design result output module is used to output the final microphone array design result.

Images