Problem with xDAWN spatial filter

Working with OpenViBE signal processing scenarios and doing scenario/BCI design
Post Reply
Luciano
Posts: 2
Joined: Wed Aug 30, 2017 7:14 pm

Problem with xDAWN spatial filter

Post by Luciano »

Hi!!!

I'm using the Cognionics HD 72 to acquisition with 8 electrodes, but I do not manage to train the filter; I obtain a warning saying :
" <Box algorithm::xDAWN Spatial Filter Trainer> Generalized eigen vector decomposition failed..."

I converted the archive (ov2csv) to see the columns.

Which column I need to delete?

https://imgur.com/a/LWsY0


Thanks!!!

jtlindgren
Posts: 775
Joined: Tue Dec 04, 2012 3:53 pm
Location: INRIA Rennes, FRANCE

Re: Problem with xDAWN spatial filter

Post by jtlindgren »

Hi Luciano,

not sure why it fails (isn't there any more information?) but I'd keep the EEG channels only. Also, the data scale seems to be massive (e+15). Check that this is as intended and then try to scale the data to some more meaningful range (e.g. simple dsp). If its not intended, check what makes it so and fix. A typical basic thing also to do is to use Signal Display to check that the signal looks like EEG in the first place. :)


Cheers,
Jussi

Post Reply