The Comparison of Room Impulse Response Measuring Systems

This paper describes cross-correlation based systems for the estimation of the room impulse response. A various types of excitation signals are used: white noise, pink noise, swept-sine, MLS, and random phase multisine. Criteria to evaluate the quality of the impulse response estimation were: a) random and stationary noise rejection, b) rejection of nonlinearities introduced by measuring system, c) immunity to time-variant environmental disturbances and d) averaging time necessary to achieve reliable estimation. The work shows that choice of an optimal excitation signal depends on measurement environment. In an environment with high noise, a random phase multisine gave the best estimation, otherwise excellent results were obtained with a log-sweep excitation. The robustness of the estimation can only be achieved with a random phase signals.