How to call this measure. (Pawel Kusmierek )

Subject: How to call this measure.
From:    Pawel Kusmierek  <pq(at)NENCKI.GOV.PL>
Date:    Fri, 9 Aug 2002 14:08:30 +0200

Dear list, I had to compare spectra produced by three different loudspeakers. I played white noise through the speakers, recorded the outpput to computer and made 1/3 octave spectra. In order to compare them quantitatively, I calculated a measure which decribes whether the spectrum goes monotonically up from lowest band to a maximum and then down to the highest band, or it wanders up and down (has troughs and peaks). Now I do not know how to call the measure. My best idea so far is 'spectrum irregularity'. Can anyone suggest a good name? Here is desription how it is calculated (from the manuscript): Moreover, [NAME] measure of the spectra was calculated as the mean of the absolute differences between neighbouring 1/3-octave bands. This measure was then normalised by dividing by the sum of ascending and descending vertical (sound pressure level) ranges of the spectra. The ranges were computed by subtracting the value in the first band and the value in the last band, respectively, from the spectrum maximum. Best regards, Pawel ================================= Pawel Kusmierek Department of Neurophysiology Nencki Institute of Experimental Biology 3, Pasteur St., 02-093 Warsaw, Poland tel. (48-22) 659 85 71 ex 379 fax (48-22) 822 53 42 E-mail: pq(at) or: kusmierek(at), pq(at) ICQ 11740175

This message came from the mail archive
maintained by:
DAn Ellis <>
Electrical Engineering Dept., Columbia University