Correction: Should read: Chebyshev I (not II) and odd-ordered to provide unity gain in the passband (within Apass dB) and odd-ordered to produce no shift in mean (therefore, if I recall, 0dB gain at zero Hz) as documented in wvDecimate. This code was written 10 years ago (so applies to old MATLAB versions) and info on MATLAB's filter implementation may be out-of-date: There's a reference to TMW Technical Solution 1-2YWEQD but that seems no longer to be available.
Corretion: Should read: Chebyshev I (not II) and odd-ordered to provide unity gain in the passband (within Apass dB) and odd-ordered to produce no shift in mean (therefore, if I recall, 0dB gain at zero Hz) as documented in wvDecimate. This code was written 10 years ago (so applies to old MATLAB versions) and info on MATLAB's filter implementation may be out-of-date: There's a reference to TMW Technical Solution 1-2YWEQD but that seems no longer to be available.
Thank you so much for you answer! E
OK. The best thing is always to reduce noise at source. That's not always easy in a theatre but the Digitmer HumBug device might be your next best option if collecting more data. It removes 50Hz and harmonics. If you have Spike2 smr files, you can remove 50Hz subtractively using a script in Spike2 written by CED. Drop them an email - they are always very helpful. It won't be as good as the HumBug which uses fast sampling real-time processing. You need an analgue filter to avoid aliasing before sampling....
Hi, first of all thank you for the fast answer! Maybe i will give you a little bit of background to better understand our problem. So, our data is collected from patients (Parkisonian patients before during brain surgery) and represents local field potential in specific brain areas and for a specific experimental paridigm the doctor asked us to perform a spectral analysis on the single channel recorded during this operation. We usually perform similar analysis on EEG signals using the EEGLAB toolbox....
Elisa Notch filters are pretty horrible things, whether analogue or digital, and introduce all sorts of distortions. You can design one in MATLAB using the standard Mathworks GUIs (fdatool or filterDesigner at the ML prompt) where they are called 'bandstop' filters. Distortion will increase with higher attentuation and narrow notches. Your data collection software provider may provide an option to subtract rather than filter away mains interference (e.g. CED). Better still, hardware to do that is...
Hi, we are trying to remove the 50Hz noise coming from the electrical line (notch filter). We are using the GUI interface and are not familiar with the matlab code. We found the digital filtering option, but we don't know how to set the parameters. Can someone help me with this issue? Thanks, Elisa
I am aware that sigTOOL does not work well as a GUI with MATLAB releases since R2012a....
I am aware that sigTOOL does not work well as a GUI with MATLAB releases since R2012a....
Not sure here which "GUI" you mean. Import in sigTOOL creates a .kcl file. That is...
Hi, I hope you can help me. I've got Alpha Omega recordings (.map format) and I'm...
For a fix for problems with unresponsive menus in R2014b and later see here.
This is due to a change behaviour from R2014b onwards. In TMW docs: Starting in R2014b,...
I cannot get the dropdown menus to do anything in 2015b. The drop-down part works,...