1. We are adding our own processing to ActiView. Normally we have been processing the raw signals at the output of Downsample-sinc.vi. Everything is fine. Now we want to process the signals after they have been referenced and filtered, just like they are displayed on the screen. So, we looked at the signals at the output of Ref-Mon-24new.vi. Not fine. These signals contain a serious glitch (spike) which corresponds to the start of the screen display. This occurs even with no reference selected. What is the cause of this glitch, and can we get rid of it?
2. Looking at the code for your filtering (FilterNew.vi) it appears that you use one filter and apply it to each channel in turn (while saving the state of course). Is this correct? Presumably you do this because you don't know in advance what channels are selected for filtering. Is that correct?
ActiView Modifications
-
Coen
- Site Admin
- Posts: 1125
- Joined: Fri Mar 26, 2004 7:00 pm
- Location: Amsterdam, Netherlands
- Contact:
The step in the signal is generated on purpose to "null" each trace at the start of the screen. This ensures that signals do not drift off screen even if they are displayed without any high-pass filtering (as pure DC signals). The signals are nulled at the start of the acquisition, every time a filter setting changes, when the "Reset Baseline" button is pressed, and at every screen start.
To get rid of the glitches, there are two options:
1) Simply disable the reset terminals ("Initialize filters" and "center traces"). This will work fine so long as you make sure that the displayed signals are always high-pass filtered (disable the option to switch off the HP filter)
2) Copy the filter loop, and remove the (two) reset cases to achieve a separate output array for your further analysis. This option will take a bit of extra processor power (extra filters applied), but it leaves the original program with screen reset intact
Your explanation of the filter setup is correct. Using a for-loop to for the filters is a convenient way to apply filtering to a variable array of signals. It is basically a variation on the LabVIEW example given in IIR Filtering and Response.vi --> IIR Filtering and Response.vi --> Implement IIR Cascade Filter for N specs and N Chan.vi
Best regards, Coen (BioSemi)
To get rid of the glitches, there are two options:
1) Simply disable the reset terminals ("Initialize filters" and "center traces"). This will work fine so long as you make sure that the displayed signals are always high-pass filtered (disable the option to switch off the HP filter)
2) Copy the filter loop, and remove the (two) reset cases to achieve a separate output array for your further analysis. This option will take a bit of extra processor power (extra filters applied), but it leaves the original program with screen reset intact
Your explanation of the filter setup is correct. Using a for-loop to for the filters is a convenient way to apply filtering to a variable array of signals. It is basically a variation on the LabVIEW example given in IIR Filtering and Response.vi --> IIR Filtering and Response.vi --> Implement IIR Cascade Filter for N specs and N Chan.vi
Best regards, Coen (BioSemi)