Re: increased range of a remote

[John Denker <jsd@AV8N.COM>]

John Mallinckrodt wrote:
> ... I find, however, that cupping my left hand around the back edge
> of my computer usually increases the signal strength by "one ring"

Yeah, there are lots of stories along those lines.  The
physics is not very easy to figure out.  First, my version
of the story:

Where my laptop is now, the "typical" WiFi signal-to-noise
ratio is about 10 or 11 dB, with wild fluctuations.  If I
move the machine a little, I can make things 3dB worse;  if
I move it a little more, I can make things 3dB better, but
it won't stay better very long, and it doesn't matter
anyway because the machine needs a steady supply of >=15dB
SNR in order to work properly.

So ... I got a piece of aluminum foil about 40cm on a side
and covered a section of desktop with it.  On top of that
there is an old book wrapped in foil, and on top of that
is the computer.  The effect is to have the wireless PCMCIA
card sticking out in the air, about 4cm above the ground
plane.  The result is 17 (sometimes 18) dB SNR ... with
improved robustness against perturbations such as exact
placement of the computer and people walking around in
the room.

I tried wearing a tinfoil hat but it didn't help. :-)

Obviously this is not a scientifically designed antenna.
I don't pretend to understand more than a small percentage
of what's going on here.  My working hypothesis is that
this location suffers from what I call "speckle".  Have
you ever defocused a laser and shined it on the wall?
You see speckle, from lots of contructive and destructive
interference.

Electrical engineers sometimes call this "multiplicative
noise".  It behaves very differently from the additive
noise that gets the lion's share of the attention in
introductory courses.

Some evidence in favor of the speckle hypothesis is that
I can _raise_ the SNR by a few dB by using an _absorber_.
In this frequency regime (ISM band, 2.412GHz to be precise)
water is pretty much just an absorber.  I observe that an
artfully placed glass of water can sometimes increase the
SNR by two or three dB.  My interpretation of this is that
the prior signal suffered from the destructive interference
of two or more paths, and that if I attentuate one of the
paths things get better.