Index of content:
Volume 118, Issue 2, August 2005
- ACOUSTIC SIGNAL PROCESSING 
118(2005); http://dx.doi.org/10.1121/1.1953307View Description Hide Description
Experimental results are presented on propagation, coherence, and time-delay estimation (TDE) from a microphone array in an outdoor aeroacoustic environment. The primary goal is to understand the achievable accuracy of acoustic TDE using low-cost, commercial off-the-shelf (COTS) speakers and microphones. In addition, through the use of modulated pseudo-noise sequences, the experiment seeks to provide an empirical understanding of the effects of center frequency, bandwidth, and signal duration on TDE effectiveness and compares this to the theoretical expectations established by the Weiss-Weinstein lower bound. Finally, sensor network self-localization is performed using a maximum likelihood estimator and the time-delay estimates. Experimental network localization error is presented as a function of the acoustic calibration signal parameters.
118(2005); http://dx.doi.org/10.1121/1.1945469View Description Hide Description
The goal of planar nearfield acoustical holography (NAH) is to recover the sound field at the sound source from pressure measurements made close to the source plane. The theory requires the pressure to be measured over a complete plane. Because experimentation consists of acquiring only a finite measurement aperture of the pressure field, it naturally causes erroneous values in the reconstructed field. Wavelet preprocessing applied to the pressure measurements in the nearfield provides a solution to lessen the effects due to the truncation of the hologram. The approach is based on a multiresolution analysis of the field from different wave number bands followed by selective spatial filtering of effects highlighted by the first analysis. Experimental results show the relevance of the method by comparison to standard NAH involving exponential filtering in the wave number domain. The computation of objective indicators based on distance measurements between wave number spectra and comparisons between patterns composed of relevant features drawn from experimental data are proposed to give objective criteria to prove the viability of the method.