IIR to find minimum, maximum enveloppe of a signal
in [DSP]
|
Prev: Tomorrow they are ready to start the LHC collider. The Earth can be exploded in a 1000 seconds!
Next: Tomorrow they are ready to start the LHC collider. The Earthcan be exploded in a 1000 seconds!
From: suren on 26 Feb 2010 11:34 On Feb 26, 4:56 pm, vectorizor <vectori...(a)googlemail.com> wrote: > > Before going off on a wild goose chase, define /precisely/ what you mean > > by "envelope". You can't compute what you can't define. > > Here is an image that shows what I want to achieve:http://img8.imageshack..us/img8/8839/11635847.png > > the signal is in blue, and the maximum and and minimum enveloppes are > respectively in red and black. FYI, I generated the enveloppes with a > method that is terribly inefficient, but it achieves good results in > terms of quality. > > Note the maximum enveloppe may be well over the signal, this is > because these are 1D projections of a 2D signal. Local extrema may be > close to the samples present in the plots, without being seen on the > plots. > > Thanks for the help. Hi, I beleive that the fast attack slow decay filter would definitely help in this case. Can you send the image values in a text file. I can look at the fast attack slow decay filter that i talked about. regards suren
From: Rob Gaddi on 26 Feb 2010 12:20 On Fri, 26 Feb 2010 03:56:38 -0800 (PST) vectorizor <vectorizor(a)googlemail.com> wrote: > > Before going off on a wild goose chase, define /precisely/ what you > > mean by "envelope". You can't compute what you can't define. > > Here is an image that shows what I want to achieve: > http://img8.imageshack.us/img8/8839/11635847.png > > the signal is in blue, and the maximum and and minimum enveloppes are > respectively in red and black. FYI, I generated the enveloppes with a > method that is terribly inefficient, but it achieves good results in > terms of quality. > > Note the maximum enveloppe may be well over the signal, this is > because these are 1D projections of a 2D signal. Local extrema may be > close to the samples present in the plots, without being seen on the > plots. > > Thanks for the help. Alright, so you're looking for a non-causal peak detector, where the effects of the peak ripple out in both directions. I _think_, and some of the other gurus here may be more smarter than I am, that you can just get that by running a simple peak detector with a 1st order IIR decay (series diode to a shunt RC) in one direction, then reverse the image and run it through in the other direction. Should give you nice symmetric peaking. -- Rob Gaddi, Highland Technology Email address is currently out of order
From: Jerry Avins on 26 Feb 2010 13:01 vectorizor wrote: >> Before going off on a wild goose chase, define /precisely/ what you mean >> by "envelope". You can't compute what you can't define. > > Here is an image that shows what I want to achieve: > http://img8.imageshack.us/img8/8839/11635847.png > > the signal is in blue, and the maximum and and minimum enveloppes are > respectively in red and black. FYI, I generated the enveloppes with a > method that is terribly inefficient, but it achieves good results in > terms of quality. > > Note the maximum enveloppe may be well over the signal, this is > because these are 1D projections of a 2D signal. Local extrema may be > close to the samples present in the plots, without being seen on the > plots. I'm glad you know what you mean. Maybe if I did too, I could offer a suggestion. In what sense do the curves represent envelopes? The strict definition of "envelope" that I know is a curve that touches every member of a family of curves tangentially. We extend this to include a low-frequency curve that touches every cycle of a high-frequency curve, as in AM. How do you define it here? Jerry -- Leopold Kronecker on mathematics: God created the integers, all else is the work of man. ����������������������������������������������������������������������
From: Michael Plante on 26 Feb 2010 17:10
Jerry wrote: >vectorizor wrote: >>> Before going off on a wild goose chase, define /precisely/ what you mean >>> by "envelope". You can't compute what you can't define. >> >> Here is an image that shows what I want to achieve: >> http://img8.imageshack.us/img8/8839/11635847.png >> >> the signal is in blue, and the maximum and and minimum enveloppes are >> respectively in red and black. FYI, I generated the enveloppes with a >> method that is terribly inefficient, but it achieves good results in >> terms of quality. >> >> Note the maximum enveloppe may be well over the signal, this is >> because these are 1D projections of a 2D signal. Local extrema may be >> close to the samples present in the plots, without being seen on the >> plots. > >I'm glad you know what you mean. Maybe if I did too, I could offer a >suggestion. In what sense do the curves represent envelopes? > >The strict definition of "envelope" that I know is a curve that touches >every member of a family of curves tangentially. We extend this to >include a low-frequency curve that touches every cycle of a >high-frequency curve, as in AM. How do you define it here? vectorizor: to clarify why a lack of a definition is a problem, picture a 1-D curve that is almost sinusoidal, but has a small "bump/peak" at the base of every cycle. Should the "max" side of "the envelope" dip all the way down just to touch a feature that you may not be interested in? And that's really not a yes/no question: it needs to be quantified. |