摘要:

This paper is concerned with the detection of a narrow‐band signal using two hydrophones. Specifically, it addresses itself to the problem of finding the “optimal” linear combination of sample spectra, cospectrum, and quadrature spectrum to use as a detection test criterion. The “optimal” detector is defined here to be the one that maximizes the average probability of detection for given probability of false alarm and given observation time, where the average is taken over all possible noise‐phase difference conditions. It is shown that under not unrealistic conditions, the “optimal” detector, among all those that are linear in the sample spectra, co‐, and quadspectrum, is the one that adds the outputs of the detectors and squares the result. However, if one is willing to consider detectors that are not linear combinations of the various spectra, there are detectors that offer significant improvements over the sum and square one and that require no additional information. [Work supported by the Bureau of Ships, U. S. Navy.]

ISSN:0001-4966    

DOI:10.1121/1.1942898

出版商:Acoustical Society of America    

年代:1966

数据来源: AIP

摘要:

In the situation where the unwanted scatterers are uniformly distributed in range and have a Doppler distribution that is independent of range, the reverberation power spectrum is completely determined by the energy spectrum of the signal transmitted. The optimum detector of an echo in noise plus reverberation is the colored‐noise type of matched filter the output signal‐to‐noise ratio of which depends only upon the energy spectrum of the echo and the power spectra of the noise plus reverberation. It is significant that this output does not depend upon the phase spectrum of the signal and hence upon its waveform. This limitation of signal design to energy spectral design does not take into account the possible effects of the angular distribution of the unwanted scatterers. The optimal energy spectrum design of signals is discussed for low and high Doppler targets, indicating the limiting values of the output of the colored‐noise filter as a function of available bandwidth. The special cases of the Urkowitz filter and MTI signals and processing are discussed.

ISSN:0001-4966    

DOI:10.1121/1.1942901

出版商:Acoustical Society of America    

年代:1966

数据来源: AIP

摘要:

There exist methods for computation of a focus that require a computer. These are based on ray acoustics in a horizontally stratified atmosphere. However, often a quick judgment must be made relative to the focal area of a high‐intensity sound source, and quick estimates are desirable. It is then adequate to determine the existence of a focus and its approximate location. The algorithm discussed here results in a simple technique for predictivg the existence and location of a focus for a common type of effective sound‐velocity profile.

ISSN:0001-4966    

DOI:10.1121/1.1942910

出版商:Acoustical Society of America    

年代:1966

数据来源: AIP

摘要:

Data are available on the surface motions at the epicenter caused by the main shock of the earthquake. These show that the permanent vertical displacements were of the order of 1 m, over an area of about 250 000 sq km. The vertical motions occurred in a period of time short as in comparison to the time for sound propagation over the strained area. The earth's surface thus became a source for an impulsive infrasonic wave in the atmosphere. The radiated‐sound field has been computed from the known surface motions, and the total radiated energy has been estimated. The results depend substantially on the structure of the atmosphere, whose scale height is much less than the linear extent of the sound source.

ISSN:0001-4966    

DOI:10.1121/1.1942912

出版商:Acoustical Society of America    

年代:1966

数据来源: AIP

摘要:

The intensity relations within a two‐tone complex were varied by setting one tone 0.5 to 4 dB below 70 dB SPL and the other the same number of decibels above. 25 subjects, listening monaurally through earphones, tried to distinguish these complexes from complexes with both tones at 70 dB. The main variable was the frequency separation ΔFbetween the two tones. It was usually more difficult to detect an intensity difference within the complex when the two tones were close or far apart in frequency than when they were at some intermediate distance, roughly equal to the critical bandwidth. The effect of ΔFwas clearest when the lower‐frequency component was more intense than the higher‐frequency component. Measures were also made with the intensity of the two components changed in the same rather than opposite directions. Here detectibility was independent of ΔF. These findings held up best at center frequencies of 1000 and 2000 cps, less well at center frequencies of 500 and 4000 cps. [Research supported by the Public Health Service, U. S. Department of Health, Education, and Welfare.]

ISSN:0001-4966    

DOI:10.1121/1.1942918

出版商:Acoustical Society of America    

年代:1966

数据来源: AIP

摘要:

Twenty subjects, both male and female, were exposed monaurally to Gaussian noise at intensity levels of 100, 110, and 120 dB SPL for 3‐min duration. Békésy‐type threshold tracings at 2000, 4000, 6000, and 8000 cps were determined for each subject and a comparison between pre‐ and post‐fatigue thresholds constituted a measure of temporary threshold shift (TTS). Each subject participated under conditions of (a) listening to the noise in one ear, and counting the “breaks” in the noise field during the 3‐min test period; and (b) hearing noise in the one ear but also listening to a story being told in the opposite ear, with the knowledge that each would be tested on the passage following the 3‐min period. The amount of auditory fatigue was shown to vary as a result of “focusing” in the opposite ear at the 110‐db‐SPL condition. It was shown that more TTS resulted in the test ear when the subjects were concentrating on the message being presented in the opposite ear.

ISSN:0001-4966    

DOI:10.1121/1.1942923

出版商:Acoustical Society of America    

年代:1966

数据来源: AIP

摘要:

Work was undertaken to develop a microphone that would provide highly intelligible speech reception by direct contact with the head. Since investigation (Paper 7D4) showed that the forehead provided the bighest quality speech, efforts were concentrated in this region. In order to avoid a loss of high‐frequency response owing to flesh compliance, the total mass of the transducer had to be limited to below 1 g for a unit of small size. This was most readily accomplished with a miniature, inertial‐type transducer, which, additionally, provides the proper response‐frequency characteristics. Articulation tests, with the microphone located on the forehead, indicated an average PB word score of 90% in quiet surroundings. Good rejection of ambient acoustical noise was observed. The ultimate noise rejection is limited by noise‐induced vibrations transmitted through the head. The microphone totally frees the mouth area of obstacles located in front of the lips, and its miniature size suggests several applications. [Work supported by the U. S. Army Electronics Command.]

ISSN:0001-4966    

DOI:10.1121/1.1942931

出版商:Acoustical Society of America    

年代:1966

数据来源: AIP

摘要:

An improved form of spectrograph utilizing a contour representation of amplitude is described. Detailed consideration is given to the novel features incorporated in the instrument such as variable sample length, fully adjustable frequency‐scan limits, and a quasistationary display. The use of contours to extend the dynamic range of the display medium is described, together with a versatile digital technique for obtaining any practical amplitude contour spacing. Finally, samples of typical plots are shown, indicating the versatility and wide application of the spectral contour plotter to acoustical signal analysis.

ISSN:0001-4966    

DOI:10.1121/1.1942938

出版商:Acoustical Society of America    

年代:1966

数据来源: AIP

摘要:

A sound system in a room, consisting of single or multiple microphones and loudspeakers connected by an active electrical link, forms a closed loop that may oscillate, or howl, at frequencies for which the magnitude of the loop gain is unity or greater and the phase shift an integral multiple of 2π. This paper gives preliminary results of detailed measurements of 10017 gain and phase, from which howl points can be predicted. The number of howl points versus gain, and the effects of air temperature, room absorption, and transducer configuration on the feedback frequencies are considered. The results have a practical bearing upon techniques of sound‐system equalization.

ISSN:0001-4966    

DOI:10.1121/1.1942946

出版商:Acoustical Society of America    

年代:1966

数据来源: AIP

摘要:

An integral relationship between the time variations of the strength of a line source and the time variations of the resulting pressure wave was derived by Levi‐Civita in 1897. In general, this integral cannot be evaluated by analytical means. An analog‐computer set‐up is shown that uses Levi‐Civita's integral to simulate a line source and generate functions that describe the resulting pressure disturbances. A number of examples are given. [Research supported in part by the Weather Radar Branch, Meteorology Laboratory, U. S. Air Force Cambridge Research Laboratories, Office of Aerospace Research.]

ISSN:0001-4966    

DOI:10.1121/1.1942948

出版商:Acoustical Society of America    

年代:1966

数据来源: AIP