+1 323 268 0395 kmw@caltek.net

A couple updates on noise measurement.

Hi Kenneth,

Thank you for your upbeat comments.  Yes, I'm looking forward to getting out from
under all these computers, getting back into Quality of Life (QOL) issues, and
collaborating with you. I believe our skills and interests will complement each
other.

A couple updates on noise measurement.  One is a new, free app for iPhones and
Android smartphones that I just learned about.  Participants can go wherever they
think there is significant or interesting noise, measure it, comment on it, and this
is all timestamped and located with GPS, then uploaded into a master PC for
analysis.  Their phones can be tested and calibrated to improve accuracy.  Here's a
link to a paper on this:

https://povesham.files.wordpress.com/2018/02/participatory-soundscape-sensing-2018.pdf

I'm impressed with their idea of having people exploring the "soundscape", instead
of setting up fixed stations for longer-term measurements right off the bat.  I
think the exploratory model takes advantage of what people already know about the
"soundscapes" they experience, as well as what they learn in the course of exploring
and using the app. 

At the same time, there will be situations where setting up a fixed station to
measure noise for 24 hours or a whole week makes sense, such as next door to a
construciton site, noisy street, or a bar/nightclub.  This is something older PCs
could be repurposed for, just running off wall power.  

There may be a solution to keeping the measuring PC indoors with a mic outdoors. 
You wouldn't want to use an analog mic and run a signal wire tens of feet into the
PC's sound card, but there are USB microphones that have the sound card stuff built
in, so you have an immediate digital representation , which is far more immune to
interference and distortion.  

https://www.ebay.com/itm/153308545876

Not the greatest mic, because it's meant for speech and cuts off low frequencies
(e.g. the rumble from the bus that just went by), but that is in the circuitry and
cheap condenser mics per se actually do pretty well, and can plug straight into a
USB sound card:

https://www.ebay.com/itm/263919160744

Normally, USB cables aren't supposed to be more than 3m/10' long, but there's a way
to use cat5/RJ45 cables as USB extenders, with little transformer gizmos:

https://www.ebay.com/itm/272980934095 

Gotta experiment and see how far this actually works!  If it works as advertised,
you could have a mic up on a roof or in the front yard.   That is only a minimal
exposure to theft and vandalism, with most of the equipment indoors.  Plus the
outdoors stuff is small and inconspicuous.

One week of noise recorded as a .WAV file at 16 bit resolution (like a CD), but with
22K samples/second (half of the CD rate), single-track is about 27 gigabytes of
data, which is quite a lot to analyze, even if not huge in relation to an ordinary
hard disk (for example, I have a pile of 80GB drives that have about zero market
value).

If software and hardware could do FFT in real time and abstract things down to, say,
 100 bytes per second, a week's worth of data is only 60 MB.  That should be a lot
more manageable.  And the software could still save short .WAV clips of extreme
events, e.g. gunshots, extra-loud  mufflers, 1000 watt car woofers sort of thing.

This should be a great Linux project, because you could set it  up once, then clone
multiple hard disks from the master, just changing a few things for the particulars
of each 'puter.  Installing Windows on multiple computers takes a lot of time,
because first you install from a DVD, but then you go online for upgrades, which
takes hours per PC.  Plus you have to buy a license and activate it on each one.

-D