Minimum-phase Impulse Response Research Articles

A new method for quantifying binaural decoloration based on parametrically altering spectral modulations

The reproduction of sound in a reverberant environment leads to spectral modifications that are typically perceived as coloration. It is well known that a dichotic instead of a diotic presentation of such signals leads to reduced perception of coloration [Salomon (1995), Ph.D. Thesis, TU Delft]. In this contribution, a new method is presented for quantifying the reduction in coloration due to dichotic presentation. In this method, the first part of a Binaural Room Impulse Response, pre-dominantly responsible for the spectral envelope is extracted, and the spectral modulations are parametrically modified and reinstated on a minimum-phase Impulse Response that is convolved with pink noise and musical instruments and presented diotically to the listeners. Their task is to adaptively adjust the spectral modulations until this diotically presented stimulus sounds equally colored as a dichotically presented binaural signal. Results show that the spectral fluctuations expressed as standard deviations in decibels needs to be reduced by 1-2 dB for the diotic presentation to sound equally colored as the dichotic presentation.

Minimum-phase impulse response channels

We give a sufficient condition that insures that the impulse response of a channel is of minimum phase. The condition is that the energy of the first path be larger than the power spectral density of all subsequent paths. Establishing a lower bound to the first path energy, this condition is more likely to be met if transmitter and receiver are in line-of-sight. Our results show that the Hilbert transform may give unreliable results, if used to extract the channel phase response from frequency-resolved amplitude measurements when the first path is strongly attenuated.

Sub-bottom profiling using ocean ambient noise

Spectral factorization is shown to restore the phase of an incoherent layered sediment reflection coefficient so that its Fourier transform is the minimum phase impulse response at each angle. The method requires the reflection coefficient to be known over a range of frequencies and the grazing angles in question to be above critical. It is developed here in the context of another recently established technique for extracting the seabed’s plane wave reflection coefficient from ambient noise data measured on a moored or drifting vertical array (VLA). Thus it offers the possibility of sub-bottom profiling from a single platform with no sound source. Limitations of the phase restoration method are discussed and, using modeled data, comparisons are made between the “true” impulse response derived from the known complex reflection coefficient and the result of applying spectral factorization to the absolute value of the reflection coefficient. The method is also demonstrated on experimental reflection loss inferred from ambient noise measurements at three moored VLA sites and one VLA drift track in the Mediterranean Sea. Sub-bottom profiles (impulse response versus position) are shown for the drift track demonstrating that one can indeed survey with only a single directional receiver. The technique appears to perform well when compared with other profiling techniques and published results.

Iterative nonparametric spectrum estimation

Starting with the traditional spectrum estimation technique, a new iterative nonparametric algorithm is developed. In each iteration of the algorithm, the effect of the previous spectral estimate is removed from the original signal by inverse filtering by a minimum-phase impulse response to obtain a residual signal whose spectral characteristics are then estimated to improve the previous estimate. The algorithm in each iteration decreases the bias of the resulting spectral estimate, without increasing its final variance. Identification of the corresponding system minimum-phase finite duration impulse response is inherent in the proposed procedure. The whole procedure can be carried out with the numerical stability of FFT's, without solving any kind of equations. This algorithm is not suitable for very narrowband spectral shapes estimation from small amounts of data. Properties of the proposed estimator are derived in detail. Experimental results show that the algorithm converges in a small number of iterations, and demonstrate that the algorithm provides very accurate spectral estimates compared with parametric methods based on ARMA models.