|
Abstract:
|
New methods of measuring impulse responses based on carefully designed deterministic
signals can further improve the performance o ered by classical methods. In fact,
these methods are particularly interesting when measuring long impulse responses as
the ones analyzed in architectural acoustics. However, the e ects of background and
impulsive noise, distortion and time-variance are known rather qualitatively. For this
reason, the ISO 18233 encourages to develop a deeper understanding of the theoretical
bases of these techniques.
In this sense, this project presents an in depth analysis of two di erent methods of
measuring impulse responses: the linear convolution of sweep signals with the inverse
lter and the circular crosscorrelation of maximum length sequences (MLS) and inverse
repeated sequences (IRS).
The results of this work reveal that the sweep technique can provide signi cant
reduction of distortion compared to MLS/IRS technique but, unlike what is explained in
the literature, sweep signals cannot reject all distortion artifacts from the causal part of
the impulse response. Besides, it is proved that IRS sequences are immune to distortion
of even order. On the other hand, it is con rmed that synchronous averaging procedure
improves the SNR at the microphone position by 3 dB per doubling the number of
averages. Alternatively, it is also proved that the noise contaminating the measured
impulse response is reduced by 3 dB every time that the length of the excitation signal
is doubled. In terms of impulsive noise, the sweep technique only contaminates speci c
frequency bands of the system's impulse response, whereas the MLS/IRS technique
uniformly distributes all impulsive noise artifacts over the entire measured impulse
response. Finally, it is also shown that MLS/IRS measurements are more vulnerable
to time-varying systems than sweep measurements. |
