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
Deciphering STFT
Dhiraj
Member
02-22-2006 04:25 AM
Options
- Mark as New
- Bookmark
- Subscribe
- Mute
- Subscribe to RSS Feed
- Permalink
- Report to a Moderator
Hi,
Refer to the attached VI (Labview 8). I am making a fiber optic pressure sensor based on photonic crystal fiber for tsunami detection. As the pressure acts on the fiber, the output intensity changes, which I have simulated in the VI. Since this is a signal which varies in both time and frequency, I have used STFT spectrogram for analysis. Unfortunately, I can’t make head or tail of the output!!
Essentially, I wish to find out the wave height (eta) from the output intensity chart.
Can someone please help me?
johnsold
Knight of NI
02-22-2006 08:56 AM
Options
- Mark as New
- Bookmark
- Subscribe
- Mute
- Subscribe to RSS Feed
- Permalink
- Report to a Moderator
I looked at the Help file for the STFT Vi and found an example. The example uses an intensity graph for the output. When I copied that to your VI the output seems to make more sense.
Take a look.
Lynn
Take a look.
Lynn
02-22-2006 11:22 AM
Options
- Mark as New
- Bookmark
- Subscribe
- Mute
- Subscribe to RSS Feed
- Permalink
- Report to a Moderator
Thanks Lynn.
The intensity graph does look better in the sense that it looks less complicated. However, can you help me in correlating the arbit signal eta, that I am currently simulating, with this intensity chart?
What is finally available to me is only the output intensity from which I have to find out the height of the wave (eta in the VI) at the point where the sensor is located.
Incidentally, if you look at the eta that I am generating, at X= 7 and 8, the value is only 0.2 and 0.5 m respectively. This is the height of the tsunami at the said depth (4000m). However, if you look at the corresponding wavelength, (Arbit Signal2 in the VI) at X= 7 and 8, Y = 150 Kms!! This long wavelength is what is distinguishing a tsunami from a surface wave and it is only this tsunami wave which generates a significant dynamic pressure which the sensor is sensing. Even a freak wave with say eta = 20m will not cause any change in the pressure ( only about 10^-37)being recorded at the bottom as its wavelength will be typically only a few meters, say 500m.
Do you think I should employ some kind of a demodulation to the output intensity ( of course I don't have the demodulation toolkit!!)? would be grateful if you could assist me in finding some relation between the STFT output (or any other function) and wave height.
Thanks again.
The intensity graph does look better in the sense that it looks less complicated. However, can you help me in correlating the arbit signal eta, that I am currently simulating, with this intensity chart?
What is finally available to me is only the output intensity from which I have to find out the height of the wave (eta in the VI) at the point where the sensor is located.
Incidentally, if you look at the eta that I am generating, at X= 7 and 8, the value is only 0.2 and 0.5 m respectively. This is the height of the tsunami at the said depth (4000m). However, if you look at the corresponding wavelength, (Arbit Signal2 in the VI) at X= 7 and 8, Y = 150 Kms!! This long wavelength is what is distinguishing a tsunami from a surface wave and it is only this tsunami wave which generates a significant dynamic pressure which the sensor is sensing. Even a freak wave with say eta = 20m will not cause any change in the pressure ( only about 10^-37)being recorded at the bottom as its wavelength will be typically only a few meters, say 500m.
Do you think I should employ some kind of a demodulation to the output intensity ( of course I don't have the demodulation toolkit!!)? would be grateful if you could assist me in finding some relation between the STFT output (or any other function) and wave height.
Thanks again.
johnsold
Knight of NI
02-22-2006 02:39 PM
Options
- Mark as New
- Bookmark
- Subscribe
- Mute
- Subscribe to RSS Feed
- Permalink
- Report to a Moderator
I have read about such tsunami detectors. It sounds like an interesting and obviously important project.
I have made some changes to your simulator. These do not necessarily provide an interpretation of the signals, but show how I often approach the analysis of an unknown or poorly known signal.
I eliminated the loop and changed the Arbitrary Signal sources to create the entire waveform at one time (also setting all to the same number of samples). I replaced the Point by Point STFT VI by the standard one. I changed Charts to Graphs. I added a selector to allow easy choice of which signal to analyze.
What I suggest you do next is try different values and see whether you can decide what parameters are most important to your analysis. I do not have enough knowledge of what is meaningful to do that.
Clearly a monitoring program would use the loops and the Point by Point analysis VIs, but I think "batch mode" may be easier for figuring out what you can do.
Lynn
I have made some changes to your simulator. These do not necessarily provide an interpretation of the signals, but show how I often approach the analysis of an unknown or poorly known signal.
I eliminated the loop and changed the Arbitrary Signal sources to create the entire waveform at one time (also setting all to the same number of samples). I replaced the Point by Point STFT VI by the standard one. I changed Charts to Graphs. I added a selector to allow easy choice of which signal to analyze.
What I suggest you do next is try different values and see whether you can decide what parameters are most important to your analysis. I do not have enough knowledge of what is meaningful to do that.
Clearly a monitoring program would use the loops and the Point by Point analysis VIs, but I think "batch mode" may be easier for figuring out what you can do.
Lynn
