Space‐ensemble average of reverberation decay curves

Abstract

In the space‐ensemble average of reverberation decay curves, {〈S2(t)〉}, based upon the decay process of total sound energy in an enclosure, the irregularities and the spatial variation inherent in the spatial average {S2(t)} and the ensemble average 〈S2(t)〉 are both effectively eliminated. Moreover, synchronized averaging incorporated with {〈S2(t)〉} minimizes the distortion due to time‐varying background noise and extends the dynamic range of the resultant curve {〈S2(t)〉}NO which, as the number of averaging samples H increases, converges to a smooth linear decay with a slope peculiar to the enclosure. Identity exists between reverberation time (RT60) evaluated on {〈S2(t)〉}NO and arithmetic mean of RT60′s on the individual spatial sample curves 〈S2(t)〉NO′s. The {〈S2(t)〉}NO could also yield the spatial average of parameters for early reflections such as Definition D(%) proposed by R. Thiele, which are less influenced by the coincidence phenomenon of reflections and better reflect the true nature of the space. The {〈S2(t)〉}NO is elegantly measured by a newly developed technique called CDS which basically allows the real‐time computation of 〈S2(t)〉 on portable equipment for field measurements, and will be conveniently used for absorption coefficient measurements as well as for the examination of sound diffusion. This paper mainly refers to: (1) basic characteristics of {〈S2(t)〉}, (2) associated synchronized averaging to give {〈S2(t)〉}NO and (3) CDS for 〈S2(t)〉NO and {〈S2(t)〉}NO measurements.