Reverberation rooms are often used for measuring the sound power emitted by sources of sound. At medium and high frequencies, where the modal overlap is high, a fairly simple model based on sums of waves from random directions having random phase relations gives good predictions of the ensemble statistics of measurements in such rooms. Below the Schroeder frequency, the relative variance is much larger, particularly if the source emits a pure-tone. The established theory for this frequency range is based on ensemble statistics of modal sums and requires knowledge of mode shapes and the distribution of modal frequencies. This paper extends the far simpler random wave theory to low frequencies. The two theories are compared, and their predictions are found to compare well with experimental and numerical results.

Energy considerations are of enormous practical importance in acoustics. In “energy acoustics,” sources of noise are described in terms of the sound power they emit, the underlying assumption being that this property is independent of the particular environment where the sources are placed. However, it is well known that the sound power output of a source emitting a pure tone or a narrow band of noise actually varies significantly with its position in a reverberation room at low frequencies, and even larger variations occur between different rooms. The resulting substantial uncertainty in measurements of sound power as well as in predictions based on knowledge of sound power is one of the fundamental limitations of energy acoustics. The existing theory for this phenomenon is fairly complicated and has only been validated rather indirectly. This paper describes a far simpler theory and demonstrates that it gives predictions in excellent agreement with the established theory. The results are confirmed by experimental results as well as finite element calculations.

Many acoustical measurements, e.g., measurement of sound power and transmission loss, rely on determining the total sound energy in a reverberation room. The total energy is usually approximated by measuring the mean-square pressure ͑i.e., the potential energy density͒ at a number of discrete positions. The idea of measuring the total energy density instead of the potential energy density on the assumption that the former quantity varies less with position than the latter goes back to the 1930s. However, the phenomenon was not analyzed until the late 1970s and then only for the region of high modal overlap, and this analysis has never been published. Moreover, until fairly recently, measurement of the total sound energy density required an elaborate experimental arrangement based on finite-difference approximations using at least four amplitude and phase matched pressure microphones. With the advent of a three-dimensional particle velocity transducer, it has become somewhat easier to measure total rather than only potential energy density in a sound field. This paper examines the ensemble statistics of kinetic and total sound energy densities in reverberant enclosures theoretically, experimentally, and numerically.

This paper examines fundamental statistical properties of the active and reactive sound intensity in reverberant enclosures driven with pure tones. The existing theory for sound intensity in a diffuse sound field, which is based on Waterhouse's random wave model and therefore limited to the region of high modal overlap, is extended to the region of low modal overlap by taking account of the random fluctuations of the sound power emitted by the source that generates the sound field. The validity of the extended model is confirmed by experimental and numerical results.

When modeling a city or a secondary road to calculate a noise map, the information about the number of heavy/light vehicles and the average speed it is not always available. In this paper, a first approach to get an automatic classification of vehicles is presented. The system is based on the classification of the audio signal that a noise source produces. Some basic classifiers have been tested (k-nearest neighbours, FLD (Fischer linear discriminator) and principal components. As first approach, the aim of the job was to determine if the different classes (trucks, cars, and motorbikes) could be separable using different time and frequency characteristics: zero crossing ratios, spectral centroids, spectral rolloff, subband energies and mel frequency cepstral coefficients. The results shows that for some of the characteristics tested, the signals are separable, so a continuous traffic noise signal could be processed to get the information of the number of heavy trucks, cars, and motorbikes that passed by during the recording time. Information of a stereo recording could be used to get information of the direction of the vehicle. At this moment, combining three characteristics and FLD, errors bellow 9% can be reported.