Author(s): Antoniadou Smaro , Papadakis Nikolaos M., Stavroulakis Georgios E.
Exponential Sine Sweep (ESS) and Maximum Length Sequence (MLS) methods are often being applied for impulse response measurements in the presence of background noise of various types and levels. Scope of this study is the estimation of the effect of background noise on acoustic parameters for ESS and MLS measurements by adding artificial background noise with the use of a sound source. Impulse response measurements were performed in an acoustic space with the use of a dodecahedral loudspeaker for ESS and MLS methods for the same source and microphone positions according to ISO 3382-1:2009. Varying levels of background noise with steps of 2 dB were applied for the measurements with the use of an additional sound source. The different types of background noise that were used were white noise, tonal, narrow band and impulsive. No background noise compensation methods were applied for ESS and MLS. The effect of different levels of background noise for octave band acoustic parameter measurements were estimated for each case by the mean absolute error compared to the measurement without artificial background noise. Results indicate that in the case of white, narrow band and tonal noise, for low background noise levels the mean absolute error for the two methods are similar. However for higher background noise levels there is a greater mean absolute error for the ESS method. In the case of impulsive noise the ESS method seems to outperform the MLS method. Implications of the findings suggest the expected deviation of acoustic parameters in the presence of background noise of various types and levels. In addition the study hints the preferred method for acoustic measurements according to the background noise and levels.
Name: Ms Smaro Antoniadou