A high-precision portable sound source measurement and positioning device and method

Abstract

The present invention relates to a high-precision portable sound source measurement and positioning device and method, belonging to the field of sound source detection and positioning. A microphone linear array is installed at the bottom of a shielding housing for receiving sound source signals. One end of a data output interface is connected to the memory of a signal processing integrated module. A rechargeable battery is fixedly connected inside the upper part of the shielding housing. One end of a charging interface is connected to the rechargeable battery. A handle is fixedly connected to the top of the shielding housing. The advantages are that it is small and light and can be carried around. Amplification and filtering can eliminate ordinary environmental noise. The fractional order cumulant statistical tool can eliminate ALPHA noise that is difficult to suppress by ordinary filtering links. By performing operations on fractional order cumulants, the ALPHA additive noise in digital signals can be eliminated, improving the accuracy and efficiency of sound source positioning.

Description

Technical Field

[0001] The present invention belongs to the field of sound source detection and positioning, and particularly relates to a high-precision portable sound source measurement and positioning device and method. Background Art

[0002] Sound is one of the important elements in nature. At the same time, various social activities in people's lives are inseparable from sound. It is not only a medium for people to communicate with each other but also an important carrier for people's production and life. With the rapid development of human society, people have entered the information age. However, information mainly consists of video, audio, etc. Mobile phones, televisions, computers, etc. in life are all closely related to sound signals. The utilization of sound signals in electronic products in life mainly lies in the transmission, encoding, decoding, etc. of sound signals. However, in addition to electronic products, the utilization of sound signals also lies in fields such as earthquake relief, geological exploration, flaw detection of welded parts, etc. The utilization of sound signals in these fields is mainly a positioning problem.

[0003] Currently, the sound source positioning and detection methods at home and abroad are mainly divided into two categories. One category is to place multiple microphone arrays at different positions to measure the sound source signals, and then calculate the sound source position based on the measured data and the fixed phase and time delay information between the microphone arrays; the other category is to use a sound signal acquisition device to collect signals and then use a specific algorithm to operate on the collected signals to obtain the sound source position information. Among them, the first measurement method has the disadvantage of high cost. On the other hand, the position placement between the microphone arrays during measurement will affect the measurement accuracy, and it is difficult to determine the optimal placement position. The second measurement method mainly relies on algorithms. Even if the algorithm performance is excellent, high-precision measurement cannot be achieved when there are deviations in the sampling equipment. On the other hand, with the increasingly complex environment, a large number of impulse shock noises have emerged. Research shows that this type of noise follows the ALPHA stable distribution, so it is also called ALPHA noise. However, the existing two major methods only perform conventional noise filtering on the sampled signals and cannot eliminate the influence of ALPHA noise on the accuracy during measurement and operation. Summary of the Invention

[0004] The present invention provides a high-precision portable sound source measurement and positioning device and method to solve the problems in the prior art that it is difficult to eliminate ALPHA noise and the error caused by the position of the microphone array, and to expand the application scenarios of the sound source positioning technology by accurately positioning the position of the sound source signal.

[0005] The technical solution adopted by the present invention is that the high-precision portable sound source measurement and positioning device includes a microphone linear array, a signal processing integrated module, a data output interface, a charging interface, a rechargeable battery, a handle, and a shielding housing. The microphone linear array is installed at the bottom of the shielding housing for receiving the sound source signal. One end of the data output interface is connected to the memory of the signal processing integrated module and is installed on the surface of the shielding housing. The rechargeable battery is fixedly connected inside the upper part of the shielding housing and is used to supply power to the microphone array and the signal processing integrated module. One end of the charging interface is connected to the rechargeable battery and is installed on the surface of the shielding housing. The handle is fixedly connected to the top of the shielding housing.

[0006] The signal processing integrated module of the present invention includes an amplification and filtering module, an AD conversion module, an FPGA module, a DSP microprocessor module, and a storage module. Among them, one connected end of the amplification and filtering module is connected to the microphone linear array, and the other end is connected to the AD conversion module. The collected sound source signal is amplified and filtered. While most of the environmental noise is filtered out, the signal to be measured is amplified to facilitate the subsequent steps. The amplified and filtered analog signal is converted into a digital signal and transmitted to the FPGA module. The FPGA module is used to mark the digital signal, and perform subsequent matrix eigenvalue decomposition and parameter pairing, and then transmit it to the DSP module for large-scale operations of fractional-order cumulants.

[0007] The microphone array of the present invention is composed of 2M + 1 microphone array elements with the same structure. The element spacing is d, and they are evenly distributed on the x-axis.

[0008] The sound source measurement and positioning method of the present invention includes the following steps:

[0009] (1) Input array receives data;

[0010] (2) Construct four specific fractional-order cumulants according to the element output signals;

[0011] (3) Construct the pseudo-inverse of the cumulant matrix;

[0012] (4) Construct the direction matrix according to the pseudo-inverse of the cumulant matrix;

[0013] (5) Direction matrix eigenvalue decomposition and parameter pairing;

[0014] (6) Use the paired parameters to solve the source position parameters;

[0015] (7) Output the source position parameters: Output the position parameters obtained in (6) to the storage module.

[0016] The step (1) of the present invention for the input array to receive data specifically includes device power-on initialization, using the microphone array to sample the sound source signal in the noise environment, and the sampled signal is s k(t); then, an array output model \(Z(t)\) is constructed based on the array model. At time \(t\), the signal output by the \(m\)-th microphone element can be expressed as:

[0017]

[0018] where \(-M\leq m\leq M\), \(f\) k is the frequency of the \(k\)-th signal, \(s\) k (t) is the sampling of the \(k\)-th signal by the microphone array at time \(t\), \(n\) m (t)=n a (t)+n b (t) is the arbitrary additive noise received by the \(m\)-th element, where \(n\) a (t) is the ALPHA additive noise, \(n\) b (t) is the ordinary environmental noise, τmk is the phase difference between the \(k\)-th signal source at elements \(m\) and \(0\), and is approximately expressed as follows:

[0019]

[0020] where r k is the distance of the \(k\)-th signal source, \(\lambda\) k is the signal wavelength and the element spacing satisfies \(d\leq\lambda\) k / 4. Taking the microphone at the center as a reference and combining the sampling signals, a microphone array output model \(Z(t)\) is established and written in the following matrix form:

[0021] Z(t)=BS(t)+n(t)

[0022] where

[0023] Z(t)=[Z -M (t),…,Z0(t),…,Z M (t)] T

[0024] B=[b1(θ1,r1),…,b K (θ K ,r K )]

[0025]

[0026]

[0027] n(t)=[n -M (t),…,n0(t),…,n M (t)] T

[0028] where [·] Tdenotes matrix transpose, where S(t) is the array input signal matrix, B is the array manifold, and b k (θ k , r k ) is the direction vector formed by the k-th signal source incident on all array elements, n(t) is the matrix composed of additive noise received by all array elements caused by the environment and array structure, etc., and θ k is the azimuth angle of the k-th signal, and r k is the distance of the k-th signal.

[0029] In step (2) of the present invention, the following 4p th order cumulants are constructed according to the definition of fractional-order cumulants:

[0030]

[0031] In the formula, cum 4p (·) represents the 4p th (0 < p < 1) fractional-order cumulant, and (·) 表示p次幂算子，(·)*表示矩阵的共轭，通过硬件部分的放大滤波可消除普通环境噪声，通过分数阶累积量的运算可以消除数字信号中的ALPHA加性噪声，上式可以写成：

[0032]

[0033] 式中表示信号加权分数阶累积量切片，(·)*表示矩阵的共轭，且0≤m,n≤M，

[0034]

[0035] 式中为4pth阶左卡普托分数阶微分，Γ(·)为伽玛函数，将分数阶累积量写为矩阵形式：

[0036]

[0037] 其中Ap为M×K矩阵，其第k列元素为：

[0038]

[0039] 由于普通待测信号均为有限带宽即窄带，且对于窄带信号sk(t)≈sk(t+1)，将sk(t+1)带入得到如下4pth阶累积量:

[0040]

[0041] 化简为矩阵形式如下：

[0042]

[0043] 其中：

[0044]

[0045] 同理，定义如下4pth阶累积量：

[0046]

[0047] 表示为矩阵形式有：

[0048]

[0049] 式中(·)H表示矩阵的共轭转置，其中：

[0050]

[0051] 同理定义如下4pth阶累积量：

[0052]

[0053] 表示为矩阵形式有：

[0054]

[0055] 其中均为M×M矩阵，且其秖为K。

[0056] 本发明所述步骤(3)中，设矩阵的伪逆为则：

[0057]

[0058] 其中ρk为的第k个特征值，uk为对应特征向量，且

[0059] 本发明所述步骤(4)中，设方向矩阵为E1,E2,E3则：

[0060]

[0061]

[0062]

[0063] 根据阵列流行与方向矩阵的相关性质易知：

[0064] E1Ap＝ApΛ

[0065] E2Ap＝ApΩ

[0066] E3Ap＝ApΦ。

[0067] 本发明所述步骤(5)中，分解与参数配对包括：分别对E1,E2,E3进行特征值分解得到相应的特征矩阵T1,T2,T3特征向量由于经过了三次独立的特征分解，T1,T2,T3在列向量的排列次序上有差异，即：

[0068] T1＝T2P1＝T3P2

[0069] 其中P1,P2为K×K矩阵，对其进行参数配对以达到排列顺序相同的目的，以矩阵T1为基础，使矩阵T2,T3的各列特征向量与T1排列一致，即可达到使特征值对应的目的，以T1,T2的配对为例，具体方法如下，令：

[0070]

[0071] 设矩阵P1的第j列中，绝对值最大的元素位于第i行(i,j＝1,…,K)，则T2的第i列向量与的第j列向量相对应，亦即的第i个对角元素与的第j个对角元素相对应，这样便实现了特征值矩阵与的配对，同理可实现与的配对。

[0072] 本发明所述步骤(6)中求解方法是：根据已配对的对角矩阵即可利用下式获取信号DOA、距离的具体数值：

[0073]

[0074]

[0075] 于是即为待测声源信号的方位角和距离参数。

[0076] 本发明优点在于：本装置体积小轻便可随身携带，可随时调整接收角度方便接收；屏蔽外壳的加持可以降低信号采集过程中的干扰，可充电电池供电可方便装置的续航及维护；通过硬件部分的放大滤波可消除普通环境噪声，分数阶累积量统计工具的能够消除普通滤波环节难以抑制的ALPHA噪声，通过分数阶累积量的运算可以消除数字信号中的ALPHA加性噪声，本发明所提供的4pth分数阶累积量矩阵的构建，不仅保留了分数阶累积量的线性和半不变性，而且不改变信号的幅值、相位信息，对ALPHA噪声具有较强的鲁棒性；同时选取不同阵元的映射关系求取分数阶累积量可以增加信号空间信息利用率，从而扩大阵列孔径提高声源定位精度和效率；另外，累积量方向矩阵的特征分解与参数配对进一步提高了算法精度；FPGA模块与DSP微处理器模块的组合能够将FPGA的硬件并行优势和DSP强大的运算能力相结合最大限度提高计算效率，数据输出接口可使参数读取更加便捷。附图说明

[0077] 图1是本发明的总体结构示意图；

[0078] 图2是本发明提供的信号处理模块结构示意图；

[0079] 图3是本发明的麦克风线性阵列接收模型简图；

[0080] 图4是本发明的声源信号定位方法流程图。具体实施方式

[0081] 下面结合附图对本发明的技术方案进行详细说明。

[0082] 如图1所示，高精度便携式声源测量定位装置包括麦克风线性阵列1、信号处理集成模块2、数据输出接口3、充电接口4、充电电池5、手柄6、屏蔽外壳7，其中麦克风线性阵列1安装于屏蔽外壳7底部用于接收声源信号，数据输出接口3一端与信号处理集成模块2的存储器相连、安装于屏蔽外壳7表面，用于与电脑等设备连接即可读取声源位置信息；充电电池5固定连接在屏蔽外壳7的上方内部，用于给麦克风阵列1、信号处理集成模块2供电，充电接口4一端与可充电电池5相连并安装于屏蔽外壳7表面，用于可充电电池5的及时充电，手柄6固定连接在屏蔽外壳7的顶部。

[0083] 如图2所示，所述信号处理集成模块2包括放大、滤波模块、AD转换模块、FPGA模块、DSP微处理器模块、存储模块，其中，放大、滤波模块相连一端与麦克风线性阵列相连另一端与AD转换模块相连，对采集的声源信号进行放大和滤波，滤除大部分环境噪声的同时对待测信号进行放大方便后续步骤的进行；AD转换模块由高速AD转换芯片构成，将放大、滤波之后的模拟信号转换为数字信号传入FPGA模块，FPGA模块用于对数字信号进行标记(根据每个麦克风阵元的位置，对相应的数字信号进行标记)，以及后续的矩阵特征分解和参数配对，再传入DSP模块用于分数阶累积量的大规模运算。

[0084] 如图3所示麦克风阵列阵列由2M+1个相同结构的麦克风阵元组成，阵元间距为d，且均匀分布在x轴上，假定K个声源信号入射到上述线性阵列，其中，θk为第k个信号的方位角，rk为第k个信号的距离，在解调到中频并抽样后，第k个信号可表示为为了分析方便且不失一般性，这里假设信号位于xoy平面。

[0085] 如图4所示，本发明声源测量定位方法，包括下列步骤：

[0086] (1)输入阵列接收数据：具体包括装置上电初始化，利用麦克风阵列对噪声环境中的声源信号进行采样，采样信号为sk(t)；然后根据阵列模型构建阵列输出模型Z(t)，如图3所示，在t时刻，第m个麦克风阵元输出的信号可表示为：

[0087]

[0088] 式中-M≤m≤M，fk为第k个信号的频率，sk(t)为麦克风阵列对第k个信号在t时刻的采样，nm(t)＝na(t)+nb(t)为第m个阵元接收到的任意加性噪声，其中na(t)为ALPHA加性噪声，nb(t)为普通环境噪声，τmk为第k个信源在阵元m和0之间的相位差，近似表示如下：

[0089]

[0090] 其中rk为第k个信源的距离，λk为信号波长且阵元间距满足d≤λk / 4，以正中心的麦克风为参考结合采样信号建立麦克风阵列输出模型Z(t)，写成如下矩阵形式：

[0091] Z(t)＝BS(t)+n(t)

[0092] 式中

[0093] Z(t)＝[Z-M(t),…,Z0(t),…,ZM(t)]T

[0094] B＝[b1(θ1,r1),…,bK(θK,rK)]

[0095]

[0096]

[0097] n(t)＝[n-M(t),…,n0(t),…,nM(t)]T

[0098] 其中[·]T表示矩阵转置，其中S(t)为阵列输入信号矩阵，B为阵列流行，bk(θk,rk)为第k个信号源入射到所有阵元上构成的方向向量，n(t)为环境以及阵列结构等导致的所有阵元接收到的加性噪声组成的矩阵，θk为第k个信号的方位角，rk为第k个信号的距离；

[0099] (2)根据阵元输出信号构建四个特定分数阶累积量：根据分数阶累积量的定义构建如下4pth阶累积量：

[0100]

[0101] 式中，cum4p(·)表示4pth(0<p<1)分数阶累积量，(·)表示p次幂算子，(·)*表示矩阵的共轭，通过硬件部分的放大滤波可消除普通环境噪声，通过分数阶累积量的运算可以消除数字信号中的ALPHA加性噪声，上式可以写成：

[0102]

[0103] 式中表示信号加权分数阶累积量切片，(·)*表示矩阵的共轭，且0≤m,n≤M，

[0104]

[0105] 式中为4pth阶左卡普托分数阶微分，Γ(·)为伽玛函数，将分数阶累积量写为矩阵形式：

[0106]

[0107] 其中Ap为M×K矩阵，其第k列元素为：

[0108]

[0109] 由于普通待测信号均为有限带宽即窄带，且对于窄带信号sk(t)≈sk(t+1)，将sk(t+1)带入得到如下4pth阶累积量:

[0110]

[0111] 化简为矩阵形式如下：

[0112]

[0113] 其中：

[0114]

[0115] 同理，定义如下4pth阶累积量：

[0116]

[0117] 表示为矩阵形式有：

[0118]

[0119] 式中(·)H表示矩阵的共轭转置，其中：

[0120]

[0121] 同理定义如下4pth阶累积量：

[0122]

[0123] 表示为矩阵形式有：

[0124]

[0125] 其中均为M×M矩阵，且其秖为K；

[0126] (3)构建累积量矩阵的伪逆：设矩阵的伪逆为则：

[0127]

[0128] 其中ρk为的第k个特征值，uk为对应特征向量，且

[0129] (4)根据累积量矩阵的伪逆构建方向矩阵，设方向矩阵为E1,E2,E3则：

[0130]

[0131]

[0132]

[0133] 根据阵列流行与方向矩阵的相关性质易知：

[0134] E1Ap＝ApΛ

[0135] E2Ap＝ApΩ

[0136] E3Ap＝ApΦ

[0137] (5)方向矩阵特征值分解与参数配对：分别对E1,E2,E3进行特征值分解得到相应的特征矩阵T1,T2,T3特征向量由于经过了三次独立的特征分解，T1,T2,T3在列向量的排列次序上有差异，即：

[0138] T1＝T2P1＝T3P2

[0139] 其中P1,P2为K×K矩阵，对其进行参数配对以达到排列顺序相同的目的，以矩阵T1为基础，使矩阵T2,T3的各列特征向量与T1排列一致，即可达到使特征值对应的目的，以T1,T2的配对为例，具体方法如下，令：

[0140]

[0141] 设矩阵P1的第j列中，绝对值最大的元素位于第i行(i,j＝1,…,K)，则T2的第i列向量与的第j列向量相对应，亦即的第i个对角元素与的第j个对角元素相对应，这样便实现了特征值矩阵与的配对，同理可实现与的配对；

[0142] (6)利用配对后的参数进行信源位置参数求解：根据已配对的对角矩阵即可利用下式获取信号DOA、距离的具体数值：

[0143]

[0144]

[0145] 于是即为待测声源信号的方位角和距离参数；

[0146] (7)输出信源位置参数：将(6)中得到的位置参数输出到存储模块。

[0147] 实验例1

[0148] 利用RK1212EN音频信号发生器产生一个22Vrms(60W)的正弦窄带音频信号作为信号源，即信源数K为1，将其置于本发明提供的高精度生源定位装置近场范围内(按普通声信号波长为17米划分，近场范围小于3倍波长),此处信号发生器分别置于10米、20米、30米、40米、50米处，每个距离处进行不同角度的实测试验，放置角度分别为30度、60度、90度。采用本发明提供的方法进行不同角度以及不同距离的声源位置参数测量绝对误差见表1，表2，其中装置所采用的麦克风阵列阵元间距d为0.5厘米，阵元数M为5个。

[0149] 表1本发明方法实测角度绝对误差

[0150]

[0151] 表2本发明方法实测距离绝对误差

[0152]

[0153] 由表1知，当信源方向一定时距离太远信号衰减较为严重测量精度会降低，距离太近时受装置阴影效应的影响精度亦会降低，但总体最低角度测量相对精度在30度50米处为0.54％；由表2知，当信源距离一定时角度太偏测量精度会降低，当声源信号与测量装置的麦克风阵列相向时接收信号最强从而测量精度最高，总体最低距离测量相对精度在30度50米处为0.57％，角度、距离测量误差均能满足大部分日常测量精度需求。

Claims

1. A method for measuring and locating a sound source, characterized in that, It includes the following steps: (1) The input array receives data; (2) Construct 4p based on the output signals of the array elements th fractional order cumulant; (3) Construct the pseudo-inverse of the cumulant matrix; (4) Construct the direction matrix based on the pseudo-inverse of the cumulant matrix; (5) Eigenvalue decomposition of the direction matrix and parameter pairing; (6) Solve the source position parameters using the paired parameters; (7) Output the source position parameters.

2. The method for measuring and positioning a sound source according to claim 1, characterized in that: The specific process of the input array receiving data in step (1) includes powering on and initializing the device, sampling the sound source signal in the noise environment using the microphone array, and the sampled signal is s k (t); then constructing an array output model Z(t) according to the array model. At time t, the signal output by the m-th microphone element can be expressed as: where -M ≤ m ≤ M, f k is the frequency of the k-th signal, s k (t) is the sampling of the k-th signal by the microphone array at time t, n m (t) = n a (t) + n b (t) is the arbitrary additive noise received by the m-th array element, where n a (t) is the ALPHA additive noise, n b (t) is the ordinary environmental noise, τ mk is the phase difference between the k-th signal source at array elements m and 0, and is approximately expressed as follows: where r k is the distance of the k-th source, and λ k is the signal wavelength and the element spacing satisfies d ≤ λ k / 4. Taking the microphone at the center as a reference and combining the sampled signals, a microphone array output model Z(t) is established and written in the following matrix form: Z(t) = BS(t) + n(t) In the formula: Z(t) = [Z -M (t),..., Z0(t),..., Z M (t)] T B = [b1(θ1,r1),…,b K (θ K ,r K )] n(t) = [n -M (t),..., n0(t),..., n M (t)] T where [·] T denotes matrix transpose, where S(t) is the array input signal matrix, B is the array manifold, and b k (θ k , r k ) is the steering vector formed by the k-th signal source incident on all array elements, n(t) is the matrix composed of the additive noise received by all array elements caused by the environment and the array structure, etc., and θ k is the azimuth angle of the k-th signal, and r k is the distance of the k-th signal.

3. The method for measuring and positioning a sound source according to claim 2, characterized in that: In the step (2), the following 4p th order cumulant is constructed according to the definition of the fractional order cumulant: where, cum 4p (·) represents the 4p th fractional - order cumulant, 0 < p < 1, (·) 表示p次幂算子，(·)*表示矩阵的共轭，通过硬件部分的放大滤波可消除普通环境噪声，通过分数阶累积量的运算可以消除数字信号中的ALPHA加性噪声，上式可以写成： wherein represents the signal weighted fractional order cumulant slice, and (·) * represents the conjugate of the matrix, and 0 ≤ m, n ≤ M where is the 4p th -order left Caputo fractional derivative, Γ(·) is the gamma function, and the fractional cumulant is written in matrix form: Among them A p is an M×K matrix, and the k-th column element thereof is: Since ordinary signals to be measured are all of finite bandwidth, i.e., narrowband, and for a narrowband signal s k (t)≈s k (t + 1), substituting s k (t + 1) into we obtain the following 4p th th-order cumulants: Simplified to matrix form as follows: Where: Similarly, define the following 4p th order cumulant: Expressed in matrix form as: where (·) H denotes the conjugate transpose of a matrix, where: Similarly, define the following 4p th Order cumulant: Expressed in matrix form as: wherein are both M×M matrices, and their ranks are K.

4. A sound source measurement and localization method according to claim 3, characterized in that: In step (3), let the matrix be the pseudo-inverse of Then: where ρ k is the k-th eigenvalue of, u k is the corresponding eigenvector, and 5. A method for measuring and locating a sound source according to claim 4, characterized in that: In the step (4), assume the direction matrices are E1, E2, E3, then: According to the relevant properties of the array manifold and the direction matrix, it is easy to know that: E1A p = A p Λ E2A p = A p Ω E3A p = A p Φ.

6. The method for measuring and locating a sound source according to claim 5, characterized in that: In the step (5), the decomposition and parameter pairing include: performing eigenvalue decomposition on Ε1, Ε2, and Ε3 respectively to obtain corresponding eigenmatrices T1, T2, and T3 and eigenvectors Since three independent eigenvalue decompositions are performed, there are differences in the arrangement order of the column vectors of T1, T2, and T3, that is: T1 = T2P1 = T3P2 Where P1, P2 are K×K matrices. Parameter pairing is performed on them to achieve the purpose of the same permutation order. Based on the matrix T1, make the column eigenvectors of the matrices T2, T3 consistent with the permutation of T1, so as to achieve the purpose corresponding to the eigenvalues. Taking the pairing of T1 and T2 as an example, the specific method is as follows. Let: Let the element with the largest absolute value in the \(j\)-th column of matrix \(P1\) be located in the \(i\)-th row (\(i, j = 1, \ldots, K\)). Then the \(i\)-th column vector of \(T2\) corresponds to the \(j\)-th column vector of, that is the \(i\)-th diagonal element of corresponds to the \(j\)-th diagonal element of, thus realizing the pairing of the eigenvalue matrix and Similarly, the pairing of and can be realized.

7. A method for measuring and positioning a sound source according to claim 6, characterized in that: The solution method in step (6) is as follows: According to the paired diagonal matrix the specific values of the signal DOA and distance can be obtained using the following formula: Then They are the azimuth and distance parameters of the sound source signal to be measured.

8. An apparatus for implementing the sound source measurement and localization method according to any one of claims 1 to 7, characterized in that: It includes a microphone linear array, a signal processing integration module, a data output interface, a charging interface, a rechargeable battery, a handle, and a shielding housing. The microphone linear array is installed at the bottom of the shielding housing for receiving the sound source signal. One end of the data output interface is connected to the memory of the signal processing integration module and is installed on the surface of the shielding housing. The rechargeable battery is fixedly connected inside the upper part of the shielding housing and is used to supply power to the microphone array and the signal processing integration module. One end of the charging interface is connected to the rechargeable battery and is installed on the surface of the shielding housing. The handle is fixedly connected to the top of the shielding housing.

9. The device for implementing the sound source measurement and positioning method according to claim 8, characterized in that: The signal processing integration module includes an amplification and filtering module, an AD conversion module, an FPGA module, a DSP microprocessor module, and a storage module. Among them, one end of the connected amplification and filtering module is connected to the microphone linear array and the other end is connected to the AD conversion module, which amplifies and filters the collected sound source signal, filters out most of the environmental noise and amplifies the signal to be measured to facilitate the subsequent steps, and converts the amplified and filtered analog signal into a digital signal and transmits it to the FPGA module. The FPGA module is used to mark the digital signal and perform subsequent matrix eigenvalue decomposition and parameter pairing. The DSP module is used for large-scale operations of fractional-order cumulants.

10. The device for implementing the sound source measurement and positioning method according to claim 8, wherein: The microphone array is composed of 2M + 1 microphone elements with the same structure. The element spacing is d, and they are evenly distributed on the x-axis.

Images