Re: [buildcheapeeg] bandwith idea

From: Joerg Hansmann (info_at_jhansmann.de)
Date: 2001-06-29 17:18:31


Hi,

----- Original Message -----
From: Lovass Laszlo <prochess_at_e...>
To: <buildcheapeeg_at_yahoogroups.com>
Sent: Friday, June 29, 2001 11:11 AM
Subject: [buildcheapeeg] bandwith idea

> Hi,
>
> An idea has crossed my mind: what if using a rate of 256 samples/sec

That is in fact the sample rate of the RS232EEG01..03

> and a
> freq band DC to 100 Hz (lowpass point here)

The RS232EEGanalog03 uses a 75 Hz 5th order lowpass.
In reality you cannot use 1/2 sample frequency as the upper
cutoff-frequency of the low-pass, because it has a limited roll-off.

> and also using a reference
> channel for detecting noises (like muscle impulses, radio freq, 50/60 Hz
> buzz, etc.)

Where exactly should the reference channel be connected to ?
To a resistor ?
To the body ?

> and subtract it in the amplifier section.

You can only compensate errors, if you can measure them exactly.
So the question is: How and where do you want to measure the
error-voltages ?

> The possible advantages: getting rid of phase and amplitude distorsion,
> possible use of gamma and even higher brainwaves (cf. 35-45 Hz brainwave
> research of visual observations, etc.).
>
...
Regards,

Joerg



This archive was generated by hypermail 2.1.4 : 2002-07-27 12:28:31 BST