Turn on suggestions
Auto-suggest helps you quickly narrow down your search results by suggesting possible matches as you type.
Showing results for
Topic Options
- Subscribe to RSS Feed
- Mark Topic as New
- Mark Topic as Read
- Float this Topic for Current User
- Bookmark
- Subscribe
- Mute
- Printer Friendly Page
In labview,how to calculate the parameters of the bilinear model that simulates the generation electroencephalogram?
jingli
Member
11-13-2004 10:16 AM
Options
- Mark as New
- Bookmark
- Subscribe
- Mute
- Subscribe to RSS Feed
- Permalink
- Report to a Moderator
Dear Sir or madam:
I have collected a lot of electroencephalogram data, which looks like a continuous sine wave plus a noise signal. what i want to do now is to fit the data with the bilinear model as the attached jpg file. the problem is the parameters of the model are changing with the evolution of the signal. would someone like to tell me how to calculate the parameters in the bilinear model?
thank you very much
I have collected a lot of electroencephalogram data, which looks like a continuous sine wave plus a noise signal. what i want to do now is to fit the data with the bilinear model as the attached jpg file. the problem is the parameters of the model are changing with the evolution of the signal. would someone like to tell me how to calculate the parameters in the bilinear model?
thank you very much
DFGray
NI Employee (retired)
11-15-2004 08:20 AM
Options
- Mark as New
- Bookmark
- Subscribe
- Mute
- Subscribe to RSS Feed
- Permalink
- Report to a Moderator
I can point you in the right direction, but probably not answer your question really well. If you have the pro or full distribution of LabVIEW, you can use the curve fitting VIs to match any set of data to a theoretical curve. The curve fitting Express VI may do all you need to do (use the nonlinear option). To get the signal evolution, break your data into pieces and process each. If you need to average a bit, you can use a sliding window of your data for each analysis, moving the window less than the window width for each analysis.
There are a plethora of curve fitting techniques built into LabVIEW - matrix operations, linear and log linear fits, Levenberg-Marquardt methods, downhill simplex, etc. You will probably need to experiment a bit to get a stable
algorithm for your case. Check your literature for ways other people have done this. There may be an easy, stable method out there.
If you are unfamiliar with curve fitting techniques, or you do not have the pro or full versions of LabVIEW, I would recommend "Numercial Recipes in C" by Press et. al., published by Cambridge University Press. The chapter on Modeling of Data will get you going. The rest of the book will provide any background you need.
This was a very general answer. If you need something more specific, let me know. Be aware that this type of problem usually requires some trial and error to get right. The answers should be tightly scrutinized before being believed.
There are a plethora of curve fitting techniques built into LabVIEW - matrix operations, linear and log linear fits, Levenberg-Marquardt methods, downhill simplex, etc. You will probably need to experiment a bit to get a stable
algorithm for your case. Check your literature for ways other people have done this. There may be an easy, stable method out there.
If you are unfamiliar with curve fitting techniques, or you do not have the pro or full versions of LabVIEW, I would recommend "Numercial Recipes in C" by Press et. al., published by Cambridge University Press. The chapter on Modeling of Data will get you going. The rest of the book will provide any background you need.
This was a very general answer. If you need something more specific, let me know. Be aware that this type of problem usually requires some trial and error to get right. The answers should be tightly scrutinized before being believed.
