[FieldTrip] Beamforming analysis in frequency band

[Tzvetan Popov]

Dear Noa,


> Thank you Tzvetan!
> 
> This raised two more question regarding the source analysis:
> 
> 1. What is the difference between defining the frequency range under cfg.foi (such as cfg.foi=[30 48]) and defining if under the cfg.tapsmofrq (such as cfg.foi=[39 39] cfg.tapsmofrq=9)?
The difference is that in the resulting output of ft_freqanalysis (lets name it freq) will contain a freq.freq field containing of 19 freq bins presuming frequency resolution of 1 Hz. Each of these is computed with smofrq of plus minus 9 Hz. 
In the second case freq.freq is one frequency bin namely 39 Hz.
> 2. the cfg.tapsmofrq option only exist in the ft_freqanalysis function, and not in the ft_sourceanalysis function. in  ft_sourceanalysis the only option is cfg.frequency which, as I mention, has to be a scalar. My question is whether defining the cfg.tapsmofrq in the ft_freqanalysis is also causing the source analysis to use the range of frequencies and not the single frequency.
Yes.
best
tzvetan

> 
> Thanks a lot,
> Noa
> 
> 2015-04-16 14:53 GMT+03:00 Tzvetan Popov <tzvetan.popov at uni-konstanz.de>:
> 
> Dear Noa,
> 
> you should use frequency smoothening to achieve this. For example, a configuration of cfg.foilim = [20 20] ; and cfg.tapsmofrq = 10; will allow you to estimate sources in the 10 to 30 Hz band. 
> 
>> Hi all,
>> 
>> I'm doing source analysis in the frequency domain of a resting-state data.
>> when doing the source analysis, I want to use several frequency bands of interest, such as beta (13-30 Hz) and gamma (30-48 Hz).
>> My problem is about the cfg.frequency parameter, which must be a scalar (whereas i'm interested in a frequency band). Arbitrarily choosing a frequency in the middle of the range doesn't seem right.
>> 
>> What is the common practice when doing source analysis using this function?
> You could evaluate this tutorial:
> http://www.fieldtriptoolbox.org/tutorial/beamformer
> 
>> Does the cfg.tapsmofrq parameter is supposed to enable choosing frequency band instead of a single peak frequency?
> and some detailed explanation in a pictorial form here: 
> https://www.youtube.com/watch?v=7eS11DtbIPw
> 
> best
> tzvetan
> 
>> 
>> Any suggestions will be very much helpful!
>> 
>> Thanks,
>> Noa
>> 
>> 
>> This is the script I’m using:
>> 
>>     cfg = [];
>>     cfg.method ='mtmfft';
>>     cfg.output ='fourier';
>>     cfg.keeptapers = 'yes';
>>     cfg.foilim = [30 48];
>>     cfg.tapsmofrq = 2; 
>>     freqClosed=ft_freqanalysis(cfg,eyesClosed);
>>     
>> 
>> 
>>     cfg            = [];
>>     cfg.method='pcc';
>>     cfg.frequency  = 39; % must be a scalar
>>     cfg.lambda     = 0;
>>     cfg.vol        = vol;
>>     cfg.grid       = grid;
>>     cfg.feedback   = 'textbar';
>>     cfg.keepfilter='yes';
>>     source1         = ft_sourceanalysis(cfg, freqClosed);
>> 
>> 
>> -- 
>> My signature:
>> 
>> http://www.documentarywire.com/earthlings/
>> _______________________________________________
>> fieldtrip mailing list
>> fieldtrip at donders.ru.nl
>> http://mailman.science.ru.nl/mailman/listinfo/fieldtrip
> 
> 
> _______________________________________________
> fieldtrip mailing list
> fieldtrip at donders.ru.nl
> http://mailman.science.ru.nl/mailman/listinfo/fieldtrip
> 
> 
> 
> -- 
> מעכשיו זאת החתימה שלי.. חובה לראות
> 
> http://www.documentarywire.com/earthlings/
> _______________________________________________
> fieldtrip mailing list
> fieldtrip at donders.ru.nl
> http://mailman.science.ru.nl/mailman/listinfo/fieldtrip

-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://mailman.science.ru.nl/pipermail/fieldtrip/attachments/20150419/f4140db1/attachment-0002.html>