SigRecML

[Piotr J. Durka]

What would you say about a high-level language -- in the spirit of
SignalML -- to describe the communication with signal recording
devices?

Yes, I'm thinking mainly about EEG amplifiers and MEG machines for
now.

That should define the stream of the data, some basic commands sent
to
the amplifier like "check the electrodes impedance" -- and maybe sth
more :)

For the data stream we might use some experience from SignalML, and
maybe also ongoing work (plans) with the multiplexer using Google
Protocol Buffers by Krzysztof Kulewski and Piotr Findeisen in the
context of OpenBCI
http://bci.fuw.edu.pl/wiki/Communication_between_BCI_software_modules.

As for hardware devices, we have some close experience with 2-3 types
of
amplifiers, which should be enough to propose for starting sth.
comparable to SignalML 1.0 (currently being replaced by v. 2.0). This
first working draft could be then used in two ongoing projects:
OpenBCI,
and connecting the signalml.org application (Svarog) to the
Braintronics
amplifier, which we will have to do in these months anyway, so why
not
do it right form the very beginning ;-)

There are some more-or-less high level protocols, like the OpenEEG
(c.f.
http://bci.fuw.edu.pl/wiki/Communication_with_EEG_hardware), BCI2000
(http://bci2000.org/ -- the site seems to be down at the moment), but
none of them to my limited(!) knowledge is general or elegant enough.

Waiting for comments,
best, Piotr
--
Piotr J. Durka, Ph.D., Dr hab., http://durka.info
University of Warsaw, Faculty of Physics.
New article in Scholarpedia:
http://www.scholarpedia.org/article/Matching_pursuit