Audio Engineering Society

Chicago Section

Meeting Review, October 22, 2003

other meeting reports

10/22/03 Meeting Highlights
by Tom Miller

On October 22, Marc Marroquin of Brel and Kjr gave a presentation on the applications of acoustic holography.



Acoustic holography was first described in 1970s Acoustical Society Journal articles, and was the subject of some innovative demonstrations by Tom Burns in the 1990s. The main commercial application now is for noise source identification. The high processing speed of modern personal computers combined with large arrays of microphones has recently made acoustic holography a much more practical acoustical tool.


Mr. Marroquin first described a method known as Spatial Transformation of Sound Fields (STSF). A sound recording from a large array of microphones (often 96 of them) is made near the sound source in an anechoic environment. The multiple channels of sound are recorded directly into a computer. There the cross spectrum of the matrix of microphone signals is calculated. From this information the software can calculate the sound pressure and velocity at any point in space ahead of or behind the microphones. A reference microphone is often placed very closed to the estimated source of the sound. The signal from the reference microphone is used to help reject non-correlated sounds that are not of interest in the measurement.


Mr. Marroquin next described Non-Stationary Spatial Transformation of Sound Fields (NS-STSF). While STSF is largely a frequency domain tool, NS-STSF is largely a time domain tool. It offers the advantage of being able to measure sounds that evolve over time, such as start up noises and transients. It is also possible to remove echoes from the measurement. It is possible to observe the flow of sound energy from one part of a noise source to another. The NS-STSF is much more computationally intensive, and has only become practical in the last 5 years.


After discussing some of the theory of acoustic holography, Mr. Marroquin showed some practical measurements. Since most of the commercial application of holography has been in the automotive industry, he showed measurements of rotating tires, acoustic leakage through a truck door, and an animation of car engine noise location as a function of crank shaft position. He catered to his AES audience as well, and had prepared a set of speaker measurements. He measured the sound coming from a suitcase-style set of satellite and subwoofer speakers before coming to the meeting. The excitation signal was a sine sweep from low to high frequency. The response of the speakers could be seen not only as a function of frequency, but also as a function of the location in the room. The radiation pattern of the speakers was easy to observe, in this case revealing some unwanted radiation emanating from the subwoofer cabinet.


Acoustic holography requires a large number of well-matched microphones. The phase matching must be better than 5 degrees. The spacing of the microphones should be no more than 1/3 to of a wavelength. One simplification that can be used with STSF measurements is to sweep a line array of microphones in front of the sound source, rather than using a full rectangular grid.


Future advancements of the holography software will allow the sound field to extrapolated not only to the plane of the sound source, but to actually predict the sound coming around corners and contours of the sound source.


While acoustic holography has been used mainly for sound source identification of machinery, there may be several applications in the pro-audio industry. Some members speculated that acoustic holography could be a powerful tool for predicting the sound field of loudspeaker horns and arrays, where far-field measurements are not practical.