From: Michal Wallace (sabren_at_manifestation.com)
Date: 2002-06-13 15:58:03
Hello Everyone,
I've been reading through the archives of the buildcheapeeg
list. I'm on message 1800 of 2803, so I guess I don't have
the full story, but I feel like I've got a good idea of
what's going on here. There's been a lot of great ideas
thrown around. I'm really excited about this project, and I
look forward to talking more with you all! :)
Here's my high level understanding of the openEEG system(s):
A) Electrodes attached to the scalp can pick up the electrical
signals produced in the brain, and send them down a wire to
a circuit board.
B) The circuit turns these analog signals to digital signals
C) Optocouplers turn these electric signals into light and then
back into electricity again. This is a safety measure
to provide a gap between the circuitry attached to the
person's head and the circuitry that leads into the computer.
D) The digital signal is fed into a microcontroller (?) that
then sends information to a PC's serial port.
E) Software on the PC monitors the corresponding COM port,
and reads the data.
F) This data can then be transformed into a data structure
similar to the "Mind Mirror" device. (From what I can
understand, this is called digital signal processing and
there are various methods for doing this, such as a fast
fourier transformation or FFT)
G) This data structure can then be displayed on the screen
or otherwise processed in various ways to give the
"feedback" to the user. ("Mind Mirror" displays, brain
conctrolled pac-man, whatever)
Is that a fair assessment?
The trouble I've noticed in this project seems to be that
the hardware parts (B,C,D) require a lot of knowledge about
electronics, and the software parts (E,F,G) require a lot of
knowledge about programming. Of course, everybody has their
strong points, and hardly anyone has a lot of free
time. There are a lot of great minds on this list, but what
seems to be missing is teamwork.
Obviously, people have been working together all along, but
to me, "teamwork" means that each person uses his or her
strengths to support the team, and relies on the strengths
of the team to augment his or her own skills.
I'd really like to try and help here.
I've looked at the circuit diagrams that Joerg and Jim M and
Andreas have posted, and I can tell that a lot of thought
has gone into them. I see people writing about the virtues
of Modular circuits and Tiny circuits and keeping the price
down and whatnot. I think that's great. The trouble is, the
ability to turn the designs into physical circuits seems to
be a rare one. As far as I can tell, only 4-5 of these
machines have actually been built.
What I'm wondering on the hardware side is: how can we make
it easier to actually build the hardware?
- is it possible to make a "simple" circuit diagram
without too much effort? Something that would take
two channels and hook up to the PC and not worry
about extensibility for now? When I say simple, I mean
like something with less than 20 parts... Something
that an electronics dummy like me could put together
in a few hours with just the parts (ordered off the
net if need be) and simple tools available at radio shack?
In other words, is it possible to take one of the
existing designs, and simply remove things that
aren't needed for a bare-bones system?
- If not, then I guess the next best thing would be to
take the design that's closest to working and figure
out exactly what needs to be done, express those
issues clearly, and come up with solutions. In other
words, someone takes a "managment" role and gets people
to focus on the issues and solve the circuitry problems.
(I say this out of the understanding that even the
ModularEEG boards are not finished... Is that true?)
Then, we come up with instructions that anyone can use to:
- order the pieces
- print out the schematics from Eagle
- and, for the electronics newbies:
- locate someone capable of putting it together, OR..
- step by step instructions on how to put it together
as in: "solder this to that" spelled out step by
step in english.
Once anyone can build the circuit, it'll be a lot easier for
the software to flourish. After all, once one person has
software, he or she can share it freely with the group. The
hardware is the tricky part because it's harder to share the
innovations. That's another reason to dumb it down: if the
hardware is a breadboard with wires sticking out if it, we
can tell people: "okay, go buy this new part and move this
wire here" and continue to work on the hardware in real
time.
As far as software goes: I think between Rob Sacks and Jim
Peters, we've got the technical know-how to build a core
library that talks to the serial port and converts the
signal into a data structure. I wouldn't presume to
volunteer either of those guys to do the work, but maybe
they'd be happy to offer advice to someone who's motivated
to get a core library working.
Once we have that, we could probably bring in someone who's
good with SWIG, and wrap the library so that anyone with
even modest scripting language skills can use it.
I think if we get that far, we'd have our tipping point.
Meaning: anyone could then build simple software or easily
build the machine or have it built. And everyone could work
from the same tools to share ideas and improve the system.
Then you've got a community built around the thing, or in
other words an established market. Then it's a lot easier to
convince (or form) a company to build kits or preassembled
EEGs.
---Anyway, I hope I'm not overstepping my bounds as a complete stranger to the group, but I'm just really impressed with what I've read so far, and I think all that's really missing is someone to act as the "go-between" and help people focus on a common goal.
I think if we did this, we could have a working system in everyone's hands in 2-3 months. I know that personally, I'd be willing to put in my own time to help coordinate things and even do some grunt work when it comes to the software. What do you all think? Are you willing to form a team here?
Cheers,
- Michal http://www.sabren.net/ sabren_at_manifestation.com ------------------------------------------------------------ Switch to Cornerhost! http://www.cornerhost.com/ High Powered Hosting - With a Human Touch. :) ------------------------------------------------------------
This archive was generated by hypermail 2.1.4 : 2002-07-27 12:28:43 BST