From: peterson_at_d...
Date: 2001-02-14 03:03:14
Dear Joerg--
I am quite excited to discover that this prototype 1-channel device
already exists. You asked me a number of questions, which may exceed
my technical competence to answer, but I will give it a shot, with
the understanding that I barely know an electron from an electrode.
1. Bandwidth--the lowest freq of practical interest is low Delta,
say, 0.5 Hz. The highest useful EEG freq is not really established,
with some people talking about 50 or 60 Hz having some importance.
But in reality, most commercial equipment (e.g. Lexicor) is blind to
anything above 30 or 35 Hz, and for many kinds of training, there may
be no real need to read anything above even 25 Hz. If necessary, one
could no doubt make do with just two "notch" bands, Theta (4-8 Hz)and
Alpha (5-13).
2. Noise--probably best thought of in terms of signal/noise ratio or
reception error rate, nicht wahr? One would not want to give
incorrect feedback to the recipient more than (to pick an arbitrary
value) 5% of the time.
3. Bits resolution--Would this not be in the 6 to 7 bit range? But
perhaps I misunderstand what is being asked here.
4. Sample rate--My thoroughly uneducated opinion would suggest that
something less than 200 per second would suffice, but again I'm not
convinced I understand the issue.
5. Electrode impedance--it is generally possible to get impedance
below 20k ohms, and the traditional standard was 10k.
I hope there is enough signal here amidst my ignorant noise to be
useful to you.
Best Regards
Jim
--- In buildcheapeeg_at_yahoogroups.com, "Joerg Hansmann" <info_at_jhansmann.de> wrote:
> Dear Jim,
>
> ----- Original Message -----
> From: <peterson_at_d...>
> To: <buildcheapeeg_at_yahoogroups.com>
> Sent: Tuesday, February 06, 2001 4:53 AM
> Subject: [buildcheapeeg] New member
>
>
> ...
> > If I can be of any help in terms of specifiying what I would want
in
> > a clinical device or whatever, I'd be happy to contribute. Also,
> > does the problem become much simpler if you are willing to settle
for
> > a 1-channel device?
>
> In the brainwaves egroups I had published a 1 channel EEG
> that is connected to the PC-gameport. However it had some
> not so advantageous technical properties (that could be partially
> enhanced).
> Especially reading the gameport under Windows at 200 samples
> per second seemed to be rather difficult.
>
> > That's all you really need for addictions/PTSD
> > training.
>
> What specifications would be desirable for
>
> 1)amplifier bandwidth ?
> 2)noise ?
> 3)bits resolution ?
> 4)sample rate ?
> 5)electrode impedance measurement ?
>
>
> CU,
>
> Joerg
This archive was generated by hypermail 2.1.4 : 2002-07-27 12:28:29 BST