Efren Fernandez Grande

Beamforming with a circular microphone array for localization of environmental noise sources

It is often enough to localize environmental sources of noise from different directions in a plane. This can be accomplished with a circular microphone array, which can be designed to have practically the same resolution over 360°. The microphones can be suspended in free space or they can be mounted on a solid cylinder. This investigation examines and compares two techniques based on such arrays, the classical delay-and-sum beamforming and an alternative method called circular harmonics beamforming. The latter is based on decomposing the sound field into a series of circular harmonics. The performance of the two signal processing techniques is examined using computer simulations, and the results are validated experimentally.

Compensating for evanescent modes and estimating characteristic impedance in waveguide acoustic impedance measurements

The ear-canal acoustic impedance and reflectance are useful for assessing conductive hearing disorders and calibrating stimulus levels in situ. However, such probe-based measurements are affected by errors due to the presence of evanescent modes and incorrect estimates or assumptions regarding characteristic impedance. This paper proposes a method to compensate for evanescent modes in measurements of acoustic impedance, reflectance, and sound pressure in waveguides, as well as estimating the characteristic impedance immediately in front of the probe. This is achieved by adjusting the characteristic impedance and subtracting an acoustic inertance from the measured impedance such that the non-causality in the reflectance is minimized in the frequency domain using the Hilbert transform. The method is thus capable of estimating plane-wave quantities of the sought-for parameters by supplying only an arbitrary initial value for the characteristic impedance. From a comparison with a simulated waveguide, it is shown that this method can accurately estimate these quantities in a waveguide that is uniform at the position of the probe. Finally, it is demonstrated how evanescent modes, characteristic impedance, and the proposed methodology can affect the measured acoustic impedance and reflectance of an occluded-ear simulator.

Experimental characterization of the Green’s function in a room using sparse reconstruction principles

Measuring the Green’s function over the entire volume of a room would typically require an unfeasible number of measurements, due to requirements on spatial sampling. To alleviate the need for excessive measurements, sparse reconstruction methods can be employed, as they make it possible to reconstruct a seemingly undersampled signal. The present study proposes a method for acquiring experimentally the Green’s function in a room by measuring directly the mode shapes of the room, based on the conception that any mode can be expanded into a number of propagating waves. If the modes are described in the wavenumber domain (as a plane-wave expansion), sparse reconstruction methods can be employed, under the implicit assumption that each mode shape is represented as the superposition of a small number of plane waves. In addition, it is assumed that the medium is source-free and homogeneous. The methodology is examined numerically and verified experimentally, based on measurements in a lightly damped rectangular room.Measuring the Green’s function over the entire volume of a room would typically require an unfeasible number of measurements, due to requirements on spatial sampling. To alleviate the need for excessive measurements, sparse reconstruction methods can be employed, as they make it possible to reconstruct a seemingly undersampled signal. The present study proposes a method for acquiring experimentally the Green’s function in a room by measuring directly the mode shapes of the room, based on the conception that any mode can be expanded into a number of propagating waves. If the modes are described in the wavenumber domain (as a plane-wave expansion), sparse reconstruction methods can be employed, under the implicit assumption that each mode shape is represented as the superposition of a small number of plane waves. In addition, it is assumed that the medium is source-free and homogeneous. The methodology is examined numerically and verified experimentally, based on measurements in a lightly damped rectangular...