Cheap Electronics Dissection Project (2006-10-11)

[Dave Long]

> With a 2gb RS-MMC card I'm sure there's enough space, but with the 
> 225MHz cpu
> patience would be required.

On a 20$ box things might need to look much more like they did back 
when people were self-hosting development systems with sub-5 MHz cpus 
-- but from the patience standpoint, that might be a feature.

> Maybe stick a spike into the ground and throw out various frequencies 
> to do
> area effect detection of metals, caves, water table, fault lines...

As far as I can tell, your energies are going to be far too weak to see 
much.  Serious exploration work involves blowing stuff up, and only 
where the neighbors complain too loudly have they recently been forced 
to switch to less explosive techniques:

http://www.ut.blm.gov/stonecabin/seismicscience.htm
http://www.glossary.oilfield.slb.com/DisplayImage.cfm?ID=285

On the other hand, instruments are getting pretty sensitive.  There is 
a story about an incident in Los Angeles where the local newspaper 
called for "information about this morning's quake", only to be told by 
a sleepy grad student that, far from having a nice solution for the 
epicenter, it seemed to be
a) at negative depth
b) moving rapidly, and
c) had probably been heading for Edwards AFB.

(Pakistan turned off all their seismometers during their nuke tests.  
Of course, everyone else could still see the reflections, but maybe 
they could've suppressed some weak signals)

> FPGAs are good for reconfigurable hardware, but does that exist for 
> antennas?

There is such a thing as a Dicke switch, which very rapidly switches 
between two antennas, or an antenna and a null load.  I'd imagine 
everything needs to be tuned and matched, but maybe you could rig up a 
design so that it would be possible to switch in and out different 
elements to reconfigure the geometry.  If you can do the signal 
processing, it may be easier to use a fixed h/w array for the raw data, 
then synthesize a virtual antenna in post.

-Dave