The impulse response of a hall can be derived from the response of the hall to any broad-band signal, as long as the signal is precisely known. The MLS technique uses pseudorandom noise as a stimulus because the impulse response can be derived with Hadamard transform, using mathematics that involves only additions. The noise signal has many disadvantages, particularly for measuring occupied halls. However with modern computers, a full Fourier transform deconvolution is almost as fast as a Hadamard transform, and other broad-band stimuli are practical. In particular, sweep stimuli have a high-average power compared to peak power, are relatively fast, and are musical. This paper presents the results of testing various sweep stimuli for the effects of loudspeaker distortion. The effects of time variance in the hall-either due to natural effects or to time variant enhancement systems-is also considered. A sweep frequency stimulus specifically designed to avoid these problems is presented. The stimulus allows an occupied hall to be measured with at least three source positions in just over a minute. The stimulus is musical and not uncomfortably loud. Results from hall measurements are given.
Click to purchase paper as a non-member or login as an AES member. If your company or school subscribes to the E-Library then switch to the institutional version. If you are not an AES member and would like to subscribe to the E-Library then Join the AES!
This paper costs $33 for non-members and is free for AES members and E-Library subscribers.