Pressure Level Spectra Research Articles

Generation of subharmonic and difference frequency acoustic waves in a water-saturated sandy sediment

Generation of subharmonic and difference frequency acoustic waves in a water-saturated sandy sediment was investigated. Subharmonic frequency acoustic wave was observed to be generated due to the nonlinearity of water-saturated sandy sediment when the fundamental frequency acoustic wave exceeded a certain threshold pressure amplitude. Pressure spectrum level of the subharmonic frequency acoustic wave linearly increased as the driving acoustic pressure amplitude increased. The pressure level was 10 dB higher than the background noise level. Generation of the difference frequency acoustic wave was also observed by the collinear acoustic waves with two different fundamental frequencies. The pressure level of the difference frequency acoustic wave was 5 dB higher than the background noise level. It seems very useful to evaluate the nonlinear parameter of water-saturated sandy sediment without disturbing the sediment. Such nonlinear acoustic responses of water-saturated sandy sediment can be utilized for the diagnosis of marine gassy sandy sediment.

Aeroacoustic tonal noise prediction of open cavity flows involving feedback

A new hydrodynamic/acoustic splitting method isemployed to predict aeroacoustic tonal noise of self-sustained oscillatory flows over the open cavity at low Mach numbers. Acoustic field is computed using a sixth-order compact scheme and a fourth-order Runge–Kutta method, with acoustic sources obtained from the unsteady incompressible Navier–Stokes calculation. First, numerical accuracy of the present splitting method is assessed for the aeolian tone generated by Karman vortex shedding from a circular cylinder at Re D = 200 and M ∞ = 0.3. A direct comparison was made with solutions of direct acoustic numerical simulation (DaNS) and Curle's acoustic analogy. The fundamental mode characteristics of the cavity flows at (i) Reδ* = 850 and M ∞ = 0.077 and (ii) Reδ* = 1620 and M ∞ = 0.147 are examined by the present method, verifying the solution with the experimentally measured sound pressure level (SPL) spectra. A dual tone characteristic observed in experiment (Henderson 2000) for case (i) is also confirmed computationally by the present method.

Aeroacoustic computation of cavity flow in self-sustained oscillations

A computational aero-acoustic (CAA) method is used to predict the tonal noise generated from a cavity of automobile door seals or gaps at low flow Mach numbers (M∞=0.077 and 0.147). In the present method, the acoustically perturbed Euler equations are solved with the acoustic source term obtained from the unsteady incompressible Navier-Stokes calculations of the cavity flow in self-sustained oscillations. The aerodynamic and acoustic fields are computed for the Reynolds numbers based on the displacement thickness, Reδ*=850 and 1620 and their fundamental mode characteristics are investigated. The present method is also verified with the experimentally measured sound pressure level (SPL) spectra.

Development of a 500-Hz masking-level difference protocol for clinic use.

The purpose of this series of experiments was to develop a simple, 500-Hz masking-level difference (MLD) protocol that could be implemented easily in the clinic to assess auditory perceptual abilities using an audio compact disc. Five, 300-ms tones with 250-ms intertone intervals were embedded in 3-s bursts of 200-800 Hz noise presented at 42.2-dB pressure-spectrum level with 4-5 s interstimulus intervals. The homophasic and antiphasic conditions were interleaved with the signal-to-noise ratios decreasing in 2-dB steps. A single-interval, "yes/no" response task was used. Three experiments were performed on 24-28 listeners with normal hearing. The mean SoNo thresholds (58.1- to 59.5-dB SPL) and the mean SpiNo thresholds (45.1- to 46.0-dB SPL) produced approximately 13-dB MLDs. Experiment 3 included a SoNpi condition that had a mean threshold of 48.8-dB SPL and a 10.0-dB MLD. The mean test, retest of the SoNo and SpiNo thresholds on 15 listeners was < 0.5 dB. Over the three experiments, 95% of the listeners had SpiNo MLDs that were > or = 10 dB.

High-Intensity Acoustic Test Facility Using Hydraulics

The hour-long duration acoustic fatigue tests at levels above 190 dB overall sound pressure level (OASPL) can be achieved with the required sound pressure level (SPL) spectrum using available test facilities with a hydraulic simulator, which operates at only 3 to 10 psi. Furthermore, the hydraulic control at the low frequencies is available, whereas there is a 50-Hz frequency cutoff for acoustic test facilities. Many structures have destructive low-frequency resonant modes which would be properly excited in the hydraulic facility and not in the acoustic facility.

The relation between gap detection, loudness, and loudness growth in noise-masked normal-hearing listeners

Gap detection was measured for 50 and 1000-Hz-wide noise markers arithmetically centered on 1000 Hz. The markers were presented in a lowpass background noise having a high-frequency cutoff of 2000 Hz. The pressure spectrum level of the background noise was either 0 or 40 dB/Hz. Gap detection was determined at sensation levels (SLs) of 10, 15, 20, and 30 dB in the 0-dB/Hz background, and at SLs of 10, 15, and 20 dB in the 40-dB/Hz background. Because loudness increased more steeply in the higher level noise background, the markers in the 40-dB/Hz background were at higher loudness (and at a steeper portion of the loudness growth function) than markers in the 0-dB/Hz background, for markers matched in terms of SL. The main goal of the experiment was to determine whether gap detection would be relatively poor at steep portions of the loudness growth function due to the gap being confused with ongoing random dips. Such confusion might be particularly great when the noise bandwidth was narrow (and therefore characterized by prominent fluctuation) and the growth of loudness was steep. Gap detection was poorer for the 50-Hz-wide marker than for the 1000-Hz-wide marker. For a given marker bandwidth at a given SL, gap detection was similar whether the marker was presented in the 0-dB/Hz noise or the 40-dB/Hz noise. Supplementary conditions indicated that (1) gap detection results for 50-Hz-wide markers presented at equal SL's were also similar between a 0-dB/Hz noise background and a 55-dB/Hz noise background, and that (2) gap detection for 50- and 1000-Hz-wide markers was poorer in a 40-dB/Hz background than in a 0-dB/Hz background when markers were matched for loudness. The results generally indicated similar gap detection thresholds for markers presented at similar SLs across a relatively wide range of background noise levels. Gap detection was related more closely to the marker-to-noise ratio than to the loudness relation between markers or the steepness of loudness growth.

Auditory Filter Widths in Children and Adults

Filter widths for six groups of subjects, adults and children ages 6, 7, 8, 9, and 10, were estimated using two different psychophysical approaches. Subjects participated in an experiment with a fixed signal level (psychophysical tuning curve procedure) and in an experiment having a fixed masker level (notched noise procedure). The psychophysical tuning curves were determined with 200-Hz-wide noise bands centered at 2.3, 2.6, 3.0, 3.4, and 3.6 kHz. The 200-ms, 3-kHz signal was fixed in level at 30 dB SPL. Notched noise thresholds were determined with notch widths of (Δƒ/ƒ) 0.0, 0.2, and 0.4. The Δƒ was the distance, in hertz, from the center frequency, ƒ, to the edge of symmetrically placed bands of noise. The bands of noise were fixed at 35-dB pressure spectrum level. Results revealed no significant differences between the children and the adults on measures of Q10, ERB, and K. The conclusion of the present study was that auditory filter widths from the ages of 6 to adult were not significantly different.

Modulation Detection in Subjects With Relatively Flat Hearing Losses

Modulation detection thresholds were measured as a function of modulation frequency in 5 normal-hearing subjects and in 8 subjects with relatively flat, slight-to-moderate hearing losses. The carrier was a broadband noise that was sinusoidally amplitude modulated (SAM) in one of two observation intervals. The spectrum level of the carrier ranged from -10 to 50 dB SPL, and, for a given carrier level, modulation frequency varied from 2 to 1024 Hz. The temporal modulation transfer functions (TMTFs) were fitted very well with a simple equation describing a low-pass filter function. The TMTFs from the normal-hearing subjects were relatively independent of carrier level, although the derived time constant tended to increase slightly with decreases in carrier level, from an average value of 2.5 msec at 30 dB SPL to 6.0 msec at -10 dB SPL. In addition, sensitivity to amplitude modulation (AM) decreased by about 4 dB as the pressure spectrum level of the carrier was decreased from 0 to -10 dB SPL. The TMTFs from 7 of the 8 hearing-impaired subjects were similar to those from the normal-hearing subjects when the carriers were presented at equal SPLs, except that the derived time constants were slightly larger in the subjects with hearing impairment. When comparisons were made at comparable sensation levels (SLs), however, the TMTFs from the two groups of subjects were quantitatively similar, with the exception that at the lowest SL (20 dB), hearing-impaired subjects typically were more sensitive to AM than normal-hearing subjects, and the derived time constants from their TMTFs were somewhat smaller. These results, taken together with previously published results, suggest that a broad listening bandwidth is important for normal performance on a temporal resolution task. That the time constant from one of the hearing-impaired subjects was significantly longer than normal, regardless of whether the comparisons were made at equal SPL or equal SL, indicates that other factors can also be important.

The detection of temporal gaps as a function of frequency region and absolute noise bandwidth.

Temporal gap detection was measured as a function of absolute signal bandwidth at a low-, a mid-, and a high-frequency region in six listeners with normal hearing sensitivity. Gap detection threshold decreased monotonically with increasing stimulus bandwidth at each of the three frequency regions. Given conditions of equivalent absolute bandwidth, gap detection thresholds were not significantly different for upper cutoff frequencies ranging from 600 to 4400 Hz. A second experiment investigated gap detection thresholds at two pressure-spectrum levels, conditions typically resulting in substantially different estimates of frequency selectivity. Estimates of frequency selectivity were collected at the two levels using a notched-noise masker technique. The gap threshold-signal bandwidth functions were almost identical at pressure-spectrum levels of 70 dB and 40 dB for the two subjects in experiment II, while estimates of frequency selectivity showed poorer frequency selectivity at the 70-dB level than at 40 dB. Data from both experiments indicated that gap detection in bandlimited noise was inversely related to signal bandwidth and that gap detection did not vary significantly with changes in signal frequency over the range of 600 to 4400 Hz. Over the range of frequencies investigated, the results indicated no clear relation between gap detection for noise stimuli and peripheral auditory filtering.

The Late Auditory Evoked Potential Masking-Level Difference as a Function of Noise Level

The purpose of this study was to determine whether the masking-level difference (MLD) obtained with the late (P2) auditory evoked potentials has characteristics similar to those reported for the behavioral MLD with respect to the effects of noise level. Psychoacoustic studies have shown that the magnitude of the MLD increases as a function of the noise level. In the present study, P2 thresholds were measured in SoNo (signals and noise in phase at the ears) and S pi No (signals out of phase and noise in phase at the ears) conditions for noise levels of 0-60 dB pressure spectrum level (Lps). Subjects were 5 normal-hearing young adults. The difference in SoNo and S pi No thresholds, or the MLD, increased linearly as the noise levels increased through 60 dB Lps, whereas the behavioral MLD magnitude is reported to saturate at high noise levels. These results suggest that the late-potential MLD reflects characteristics similar to, but not identical with, those of the behavioral MLD.

Tuning curve widths of infants and adults in simultaneous and nonsimultaneous masking

Tuning curve widths were estimated using the pulsation threshold (PT) technique for infants, 6‐month‐olds, and adults. The observer‐based psychoacoustic procedure [L. W. Olsho et al., Dev. Psychol., in press (1987)] was the test procedure. Listeners discriminated a train of alternating masker and probe tone bursts from a train of masker tone bursts presented with a continuous probe tone. These stimuli were presented in a background of broadband noise at 20‐dB pressure spectrum level. The probe level was 10 dB SL. PTs were obtained at the tuning curve tip and at two points, one above and one below the probe frequency, 10 dB above the tip. The resulting Q10s were compared to those obtained in simultaneous masking for 6‐month‐olds and adults [L. W. Olsho, Inf. Behav. Dev. 8, 371–384 (1985)]. Adult Q10s were significantly greater for PT tuning curves at 500, 1000, and 4000 Hz. Infant PT Q10s were significantly greater than simultaneous masking Q10s at 500 and 1000 Hz, but not at 4000 Hz. [Work supported by NIH.]

Influence of signal duration on the masking-level difference.

The effect of signal duration on the 500-Hz masking-level difference was studied in 4 subjects using an adaptive, two-interval, forced-choice method. SoNo, S pi No, and SoN pi thresholds for 13 signal durations between 2 and 100 ms were established in continuous broadband noise (36.2 dB pressure-spectrum level). The signals were synthesized and gated digitally with 1-ms rise-fall times. The SoNo threshold integration functions diverged from the S pi No or the SoN pi threshold integration functions as the signal duration was reduced from 100 ms to 6-10 ms. As the signal duration was reduced below 6-10 ms, however, the threshold integration functions for SoNo, S pi No, and SoN pi converged. The relationships among the threshold integration functions were reflected in the SoNo-S pi No and the SoNo-SoN pi masking-level differences that increased as the signal duration was reduced from 100 to 6-10 ms and then decreased as the signal duration was reduced below 6-10 ms.

Effects of signal duration on the 500-Hz masking-level difference.

Two experiments were conducted to study the masking-level difference (SoNo-S pi No thresholds) for 2-128-ms, 500-Hz tones in 32-dB pressure-spectrum level, continuous broadband noise. In Experiment 1 (N = 8) the Békésy tracking procedure was used, whereas in Experiment 2 (N = 3) an adaptive, two-interval forced-choice (2IFC) psychophysical procedure was used. The data from both experiments indicated that as the signal duration was shortened from greater than or equal to 100 to approximately 10 ms the SoNo and S pi No threshold integration functions diverged, thereby producing an increase in the masking-level difference. For signal durations less than 10 ms, however, the SoNo and S pi No threshold functions converged and the masking-level difference decreased. In Experiment 2, the mean masking-level difference decreased significantly by 1.7 dB as the signal duration was shortened from 10 to 2 ms. This new finding, which was not anticipated, has recently been confirmed in another laboratory (Cokely & Hall, 1985).

The Binaural Masking Level Difference as a Function of Frequency, Masker Level and Masking Bandwidth in Normal-Hearing and Hearing-Impaired Listeners

Binaural masking level differences (MLDs) were measured for 10 normal-hearing and 10 cochlear-impaired listeners, at 500 and 2 000 Hz. Maskers were either wide-band (approximately 2 000 Hz), or narrow-band (50 Hz). For wide-band maskers the noise pressure spectrum level was 50 dB/Hz, and for the narrow-band maskers the noise levels were 50 and 60 dB/Hz. At 500 Hz the hearing-impaired listeners show abnormally low MLDs in wide-band noise, but not in narrow-band noise. At 2 000 Hz, the hearing-impaired subjects show abnormally low MLDs only in narrow-band noise at the 50 dB/Hz level. The clinical implications of the results are considered.

Masking-level difference in filtered-random and amplitude-modulated noise.

The masking characteristics of filtered-random noise and amplitude-modulated noise (40-dB pressure-spectrum level) were studied in a series of four experiments using a masking-level difference paradigm, that is, 500-Hz thresholds in S0N0 and SpiN0 binaural conditions. In Experiment 1 a filtered noise (200-800 Hz) and an amplitude-modulated noise (425-575 Hz) produced equal thresholds in S0N0; for SpiN0, however, the thresholds in amplitude-modulated noise were 1.6 dB lower than were the thresholds in filtered noise. In Experiment 2 S0N0 and SpiN0 thresholds were established in three filtered noise bandwidths (200-800 Hz, 335-685 Hz, and 410-600 Hz) and in the amplitude-modulated noise (425-575 Hz). When the bandwidths of the filtered-random and amplitude-modulated noises were similar, the SpiN0 thresholds were the same but the S0N0 thresholds were different. Experiments 3 and 4 indicated that with amplitude-modulated noise when the zero amplitude crossings of the tone and the noise coincided, the S0N0, SpiN0, and SmNm thresholds were 1.2-2.0 dB higher than were the thresholds when the zero crossings of the tone and noise did not coincide. The data indicate that the masking characteristics of filtered-random noise and amplitude-modulated noise are different. The findings are interpreted as an indication that perceptual phenomena depend on the temporal characteristics as well as the power spectrum of the signal and masker.

Analysis of acoustic ambient noise data for Monterey Bay, California

Analysis has been carried out of sonobuoy noise data recordings made in Monterey Bay at various times during 1980 and 1981 by Wilson, Wolf, and Ingento, Calibrations permitted reducing sound pressure spectrum levels to a standard reference level. The dependence of levels upon range from shore as a function of wave and surf conditions was investigated over the frequency from 10 to 2500 Hz. During heavy wind and surf conditions, levels are significantly higher than corresponding levels typical of the deep ocean, such as from the Wenz curves. At frequencies in the range 100–500 Hz, the decrease in levels with increasing range from shore and their increase with increasing swell wave energy support the conclusion that waves breaking at the beach can contribute significantly to the ambient noise in coastal waters. [Supported in part by funds from the Naval Research Laboratory.]

Additive masking effects of noise bands of different levels.

The simultaneous, ipsilateral masking of a sinusoid of frequency fs (600–860 Hz), located near the common cutoff frequency fc (750 Hz) of a low-pass and high-pass noise was examined as a function of the level of each of the bands. The pressure spectrum level of the high-frequency band Lh ranged from 5–50 dB re 20 μPa in 15-dB steps and the level of the low-frequency band L1 was from 0–24 dB down from each of these levels. The computer-synthesized noise provided a spectral step function transition at fc. The data generated by each of the three listeners in the 4IFC adaptive task was virtually identical. When fs≫fc masking was determined simply by Lh, and when fs≪fc masking was generally equal to but sometimes greater than that attributable solely to L1. When fs?fc, masking was intermediate relative to the asymptotic values just mentioned. The data are consistent with a conceptual auditory filter which has a Gaussian shape, an equivalent rectangular width of 110 Hz (for the limited frequency range considered), and a center frequency which is shifted away from the high-band noise relative to fs. These characteristics are not dependent on Lh although the asymmetry of filter positioning is weakly related to L1−Lh. The data are also consistent with a model which postulates that masking is a function of the simple sum of the masker power derived from the low-band, high-band, and internal noise appearing at the filter output.

A quantitative relationship between geometry reconstruction inaccuracies and errors in the SPL correction for the Lloyd's mirror effect

In attempting to calculate the spectral sound pressure level of an acoustic signal from measurements made in an underwater environment subject to surface interference effects, spectral amplitude correction techniques must be used. One such technique is based on the reconstructed source-receiver geometry. For the static situation in which the geometrical parameters are accurately known, a reliable correction can be made. For the dynamic case, however, in which there is a relative motion between the source and receiver, inaccuracies are likely to exist in determining the pertinent geometrical parameters such as source and/or receiver depths, source-receiver range, relative velocity, and sound speed. These inaccuracies translate directly into errors in the calculated sound pressure spectrum levels. The present analytical study quantitatively relates errors in determining the geometrical parameters to errors in the resulting Lloyd mirror corrections for the situation involving relative motion between the source and receiver. [Work sponsored by Naval Air Systems Command.]

Infrasonic sea noise at the Mid‐Atlantic Ridge near 37 °N

Underwater ambient noise spectra, spectrum levels, and statistics are described for data continuously recorded in the 5–32‐Hz band for 176 h. These measurements were made using two ocean bottom seismographs located in 2478 m of water within the seismically active median valley of the Mid‐Atlantic Ridge near 37 °N. Laboratory tests of the H‐38 hydrophone show that flow noise levels developed by a 0.3 knot current—the maximum speed anticipated in the median valley—did not exceed the electronic noise level of the recording instrument. Over 500 underwater earthquakes were detected and epicenters were estimated to lie within about 20 km of the instruments. Seismic disturbances raised noise levels at least 20 dB across the acoustic band, varied in local magnitude from 1.2 to 2.5, persisted for as long as 2 min, and occurred on the average of 2.9 events/h. Sound pressure spectrum levels averaged over the entire recording period and compared to shipping noise curves proposed by Wenz and Urick infer that infrasoni...

On the Energy Transported with a Sound Pulse

Definitions of Fourier pressure spectrum level and integrated band pressure level are proposed for engineering description of the spectra of sonic booms and explosive sounds, and relations to specific energy spectral density are demonstrated.