The MM-Wave Group also carry out research in Acoustic Interferometry and Imaging techniques. At first appearance, acoustics seems an odd companion to mm-waves, but the two areas have a lot in common. In each case we are dealing with wide-bandwidth signals with wavelengths of the order of millimetres to centimetres. The group uses an anechoic chamber in our Physics building. The main project so far has been to develop a novel way to measure the imaging performance of hi-fi loudspeakers.
One important aspect of a high-quality stereo system is to provide a clear, stable image of the recording. This is particularly important in recordings of classical music and jazz where each instrument should sound as if it is in a clearly define place in front of the listener. The instruments should not seem to wander about as they play different notes. Nor should the harmonics they produce come from various places.
To achieve a good image it is vital that the apparent acoustic position of each loudspeaker is essentially fixed and does not vary with the signal frequency or level. By using our interferometric techniques we are able to obtain 3D information on the speaker's apparent location and hence deterime how stable the image a pair of the speakers produces will be. Although nominally for stereo, this same work has even more importance in home theatre multi-speaker systems to ensure a convincing surround/3D sound picture.
Standard loudspeaker measurements concentrate on the basic properties of the speaker such as its frequency response. For some years people have been performing laser holography to examine vibrations of the various parts of the speaker. Now we are developing systems which can measure directly the sound imaging and spatial coherence of loudspeakers. This will lead to better loudspeakers and improved stereo systems.
Although this application is quite different to those in our mm-wave work, the basic techniques of spatial interferometry still apply. In addition we can often model complex (i.e. expensive) mm-wave systems with an acoustical equivalent. In this way we can try out various ideas for sensor layout and signal processing relatively easily using soundwaves before making a mm-wave equivalent.
Content and pages maintained by: Jim Lesurf (firstname.lastname@example.org)
using HTMLEdit3 on a StrongARM powered RISCOS machine.
University of St. Andrews, St Andrews, Fife KY16 9SS, Scotland.