Spindle peak distance hardcoded

[jens-k]

@NNiethard
Can we make these 100/125 ms dependent on sampling rate and detection frequency?

Matlab/detectEvents.m

Line 559 in ddf2f5a

if sum(double(isLongEnough))>1 && sum(double(MaxIsThere))>1 && max(diff(locs))<100 %check if long enough spindle is present and check that no peak to peak distance is more than 125ms

[jens-k]

Now its 0.125 * Fs. Can we replace the first factor by 1/cfg.spi_freq(1) ?

[NNiethard]

In general yes however this might lead to bad results if someone desided to use a very narrow filter band. Maybe we can include a warning?

[jens-k]

After skype: Let's have this as a cfg option. e.g.:

cfg.spi_maxpeakdist = 0.125; % 0 for 'off'