The problem of beamforming and related beamspace high resolution direction-of-arrival (DOA) estimation is studied in this paper. All beamspace processing methods are based on the beam outputs and the beampattern design plays an important role in providing high quality beam output data for further processing. Three typical situations which are frequently encountered in practical sonar system working environment and the most widely studied MUSIC algorithm are considered herein. First, when isotropic noise is the dominant noise at sensors, conventional beamforming techniques provide the optimum performance in the sense that DOA estimate is the ML estimate. Good DOA estimates are obtainable by applying MUSIC to the beam outputs directly. Then, uncorrelated interferes with much higher strength than the wanted signals are assumed to be present in the sidelobe region, and low sidelobe Dolph-Chebyshev and adaptive MVDR beampatterns are designed to guarantee the performance of MUSIC. And finally, the robustness of conventional techniques is combined with the adaptivity of MVDR beamforming to deal with the situation when the interfere in the sidelobe region is strongly correlated with one of the wanted sources. Performance in all three situations is studied with numerical examples.
The system to be developed in the present study will enable us to extract the tumor region automatically from three-dimensional ultrasonic images of the breast, and differentiate benign from malignant tumors by using the characteristics of their surface form. In application of such a system, the accuracy of diagnosis greatly depends on its ability to extract tumor automatically. We developed an algorithm for determination of the tumor region using fuzzy reasoning, that is, we classified each voxel of three-dimensional images as “tumor,” “normal tissue” and “boundary,” and, using relaxation techniques to resolve regional contradictions, made final decisions as to the tumor region. It must be noted that, according to this algorithm, a three-dimensional the space differentiation filtering automatically generates a membership function for the fuzzy reasoning. Previous attempts of extracting tumor used a single LoG filter as the space differentiation filter. We recently developed a method which can cope with diverse ultrasound images more flexibly. With this new method, multiple DoG filters with varying characteristics are prepared in addition to the LoG filter, and the optimum one is selected from multiple extraction results. The introduction of this method improved the accuracy of extraction.
In order to auralize the room impulse responses obtained by the numerical analysis based on the wave theory, a multi-channel sound field simulation system using the Finite Difference Time Domain (FDTD) method has been developed. In this system, uni-directional impulse responses for four orthogonal directions in a 2-dimensional space are firstly calculated by the FDTD method and they are reproduced directly from four loudspeakers set in an anechoic room. At the center point of the reproduced sound field, we can hear the sound with 2-dimensional information assumed in the FDTD calculation. In this paper, the principle and the basic performance of this system are introduced and an example of subjective hearing test performed using this system is presented.
Blind beamforming algorithms have the ability to recover the desired signals from sensor array outputs without any prior knowledge of the direction-of-arrivals (DOAs). Non-Gaussian signals with negative kurtosis can be automatically captured by the multistage constant modulus (CM) array, which is the most striking blind beamforming algorithm and has been widely discussed in literatures. However, the sources number must be pre-determined in all kinds of blind beamforming algorithms. Based on the multistage CM array, we present a new method in this paper. It is designed to recover the desired signals and automatically determine the number of sources simultaneously. If the array geometry is known, the DOAs of all sources also can be estimated at the same time. The performance of the new method was analyzed via computer simulations and water tank experiments, and compared with that of other DOA estimation algorithms including “non-blind” and “blind” ones under the assumption of knowing the sources number. The new method shows better results in all considered situations.
A high performance pitch detection algorithm, called harmonic wavelet transform method, was proposed. Since the algorithm is based on a continuous wavelet transform, the cost of computation is high. However, real-time processing of the algorithm is required for some acoustical applications, such as multi-modal interface which can take into account of human emotion. Digital Signal Processor (DSP) is suitable for implementation due to its compactness. However, implementaion of the algorithm on DSP costs too much with respect to both time and funds. In order to release the issues, one of other devices is a cluster system. The cluster system can be constructed with ease because the computer node has recently becomes inexpensive. Moreover, software packages for parallel and distributed computing can be obtained without difficulty. From the viewpoint of acoustical signal processing services on the Internet, the implementaion on network connected systems, such as the cluster system, becomes an important issue for ubiquitous and grid computing. This paper proposes the parallel algorithm of the harmonic wavelet transform method. Furthermore, the proposed algorithm is implemented on a signal processing system based on cluster system. As a result, the proposed parallel algorithm is executed in real-time due to both the proposed parallel algorithm and the constructed real-time signal processing system.
In this paper, a new theory for achieving the active suppression of reflected sound waves from the walls in a room is proposed with its basis on Kirchhoff-Helmholtz boundary integral equation and inverse system theory. By actively suppressing and absorbing the unwanted reflections in the target sound field, the proposed method controls the control sources distributed outside the control region so as to equalize the region to the free field. For this purpose, Kirchhoff-Helmholtz boundary integral equation gives an effective means. Moreover, a method for suppressing the reflected sound waves from a certain part of the boundary is also mentioned. The validity of those methods is demonstrated via computer simulations.
This paper proposes a microphone using a new principle, involving the optical reflection at the curved end surface of a thin glass rod or a glass fiber. This microphone uses the change in the optical total reflection area by the change in the refractive index by the sound pressure. The change ratio in the reflection area depends on the shape of the curved end. This paper investigates theoretically the effect of the curved end shape on the sensitivity for the spherical end and the parabola type end. The microphone is expected to have possibility of practical use, though the sensitivity is low. The sensitivity may be improved by the invention of curved end shape and the detection method. This microphone is expected to have good high frequency response as it can be made small and involves no mechanical vibration.