Feature extraction method using Mel frequency cepstrum coefficients (MFCC) based on acoustic vector sensor is researched in the paper. Signals of pressure are simulated as well as particle velocity of underwater target, and the features of underwater target using MFCC are extracted to verify the feasibility of the method. The experiment of feature extraction of two kinds of underwater targets is carried out, and these underwater targets are classified and recognized by Backpropagation (BP) neural network using fusion of multi-information. Results of the research show that MFCC, first-order differential MFCC, and second-order differential MFCC features could be used as effective features to recognize those underwater targets and the recognition rate, which using the particle velocity signal is higher than that using the pressure signal, could be improved by using fusion features.
22 July 2024 in IEEE Journal of Oceanic Engineering
We investigate the beamforming method for detecting signals and estimating their directions of arrival for an acoustic vector-sensor linear array mounted near a baffle. The array-measurement model is established based on the baffle reflection coefficient, which is calculated using transfer matrices and matched boundary conditions. Herein, we propose a conventional beamforming (CBF) algorithm based on polarization filtering. First, the CBF output is directly decomposed into a sum of polarization monochromatic signals by the quaternion Fourier transform. Next, polarization filtering is conducted on each monochromatic signal and is regarded as a postprocessor to the CBF algorithm. This filtering in the polarization domain can improve the signal-to-noise ratio of the CBF algorithm. The proposed algorithm produces a lower noise level than the CBF algorithm. Moreover, the performance of this algorithm is related to the degree of polarization of the noise, and the best performance is obtained as the degree of polarization approaches zero. In this study, the performance of the proposed algorithm is evaluated by simulations and experiments.
1 January 2024 in IEEE Transactions on Geoscience and Remote Sensing
Since it is difficult to obtain the number of sources accurately in practical situations, direction-of-arrival (DOA) estimation methods without the number of sources are very much needed for non-cooperative targets and maritime attack-defense pairings. In this paper, a DOA estimation method using acoustic vector sensors (AVS) without a known number of sources is proposed. First, a reference matrix is introduced, and the eigenvalues corresponding to the noise of this matrix converge to 0. Next, a scanning matrix containing the scanning sources is created, and the analysis of noise power invariance is used to determine the power selection method of the scanning sources. The ordering law of the eigenvalues of the source matrix scanned at different angles is applied, and the purpose of estimating DOA without the number of sources is accomplished. The proposed algorithm is capable of performing DOA estimation in situations where the number of sources is uncertain and it does not rely on the array structure, which means it can be used for arbitrary arrays. Finally, the correctness and effectiveness of the proposed algorithm are verified by simulation and experiment.
In this work, we design a small-sized bi-cone acoustic vector-sensor array (BCAVSA) and propose a frequency invariant beamforming method for the BCAVSA, inspired by the Ormia ochracea’s coupling ears and harmonic nesting. First, we design a BCAVSA using several sets of cylindrical acoustic vector-sensor arrays (AVSAs), which are used as a guide to construct the constant beamwidth beamformer. Due to the mechanical coupling system of the Ormia ochracea’s two ears, the phase and amplitude differences of acoustic signals at the bilateral tympanal membranes are magnified. To obtain a virtual BCAVSA with larger interelement distances, we then extend the coupling magnified system into the BCAVSA by deriving the expression of the coupling magnified matrix for the BCAVSA and providing the selecting method of coupled parameters for fitting the underwater signal frequency. Finally, the frequency invariant beamforming method is developed to acquire the constant beamwidth pattern in the three-dimensional plane by deriving several sets of the frequency weighted coefficients for the different cylindrical AVSAs. Simulation results show that this method achieves a narrower mainlobe width compared to the original BCAVSA. This method has lower sidelobes and a narrower mainlobe width compared to the coupling magnified bi-cone pressure sensor array.
With the development of techlology, noise controlling has received wide attention in recent years. Noise source identification is the key step for noise controlling. Spherical microphone array, which can locate the noise source of arbitrary direction in three-dimensional space, has been widely used for noise source identification in recent years. Conventional methods of locating noise source include spherical near field acoustic holography and spherical focused beamforming. The acoustic quantities are reconstructed by using spherical near field acoustic holography method to realize the noise source identification, while the noise source can also be located by using focused beamforming based on spherical harmonic wave decomposition. However, both these methods have their own limitations when they are used in identifying the noise source. Spherical near field acoustic holography has low resolution at high frequency with a far distance from noise source to measurement array for noise source identification, whereas the spherically focused beamforming has low localization resolution at low frequency. Noise source identification is discussed here, and a 64-element microphone spherical array with randomly uniform distribution of elements is designed. The combination methods of noise source identification by using spherical near field acoustic holography and mode decomposition focused beamforming are investigated. The performance of the proposed combination method is simulated, and an experiment on noise source identification is carried out based on the designed spherical microphone array to test the validity of proposed method. Research results show that the high-resolution noise source identification can be achieved by using near field acoustic holography when reconstruction frequency is 100-1000 Hz with a distance 0.3-0.45 m from noise source to the center of spherical array, while high resolution of noise source localization can be achieved by using spherical wave decomposition beamforming when signal frequency is 1000-5000 Hz with a distance 0.5-3 m from noise source to the center of spherical array. Spherical array with random uniform distribution of elements maintains stable identification ability in all bearings. The spherical near field acoustic holography has high-resolution distinguishing ability in near field and at low frequency, while the focused beamforming method has high-resolution distinguishing ability in far field and at high frequency. Therefore the noise source can be efficiently identified by using the proposed combination method of near field holography and focused beamforming with less elements and small aperture spherical microphone array.