baudline
FFT Analyzer
Amazing spectrum analysis software.
Free download exclusively for Linux!
Home
News
What is baudline?
Screenshots
Download
FAQ
Manual
Search
Solutions
Mystery Signal
Contact
Mystery Signal of the Month   Aug 24 2007
We've got signal, but what the heck is it?
That's your mission.  Analyze this bébé.
 

mystery spectrogram                                                                                                                                                          

download the audio file 17.wav.bz2

hints :

  • No need to bunzip2 this file since baudline can automatically uncompress bzip2 files.
  • Setup baudline to be a Web Browser helper application as described in the FAQ.
  • Slow down the play back speed to 0.125X or 0.250X with the Play Deck's speed control and listen for sonic details that would otherwise be missed.
  • Use a larger FFT size, a narrow Gaussian window, and the harmonic bars to determine the low frequency harmonic relationships.
  • Use the periodicity bars to measure the timing between the wideband events.
  • The high frequency bandwidth looks like it would extend past 24kHz if a higher Nyquist sampling rate was used.
  • What does the purple and green banding signify?
  • What do the decay tails at 7500 and 12500 Hz signify?



What is it?

This mystery signal is the mechanical double click sound of a 3-button Logitech optical wheel mouse .  The file has two distinct double click events while only one event is visible in the above spectrogram.

Two microphones, one on each of the left and right sides of the mouse, recorded the clicks in a close stereo layout.  The purple and green banding at first looks like a natural phasing effect but further observation shows that the even and odd impulses share a unique signature.

A Gaussian window with a beta of 5.0 was used to improve the time resolution of the spectrogram image.  This allowed the observation of two useful features; periodicity and ringing.

The 4 impulses appear to be evenly spaced quadruplets but careful use of baudline's periodicity bars shows that there is a large amount of variation between all the impulse.  The delta spacing ranges from 75 to 98 ms.

The tightening of the broadband impulse also allowed the ringing decay tails to be visible.  The decay tails at 7500 Hz only occurs with the first and third impulses while the 12500 Hz tails occur with all 4 impulses.  This even-odd ringing behavior matches the color banding mentioned above.  What is happening is that the button down press allows part of the metal switch structure to ring at 7500 Hz while the button up release action mechanically damps the ringing. 

The harmonic helper bars show that the ringing at 450, 630, 2100, 7500, and 12500 Hz is not harmonically related.  The frequency structure of ringing bells is similar so this isn't that surprising.

Copyright © 2008 SigBlips.com - group - blog - site map