envelope and demodulation analysis

[rpursley8]

That works very well if the original signal is DC shifted before it is modulated (as it is for AM radio), but for a general application the modulated signal and how it is modulated are unknowns you cannot just use the simple envelope detector.  Use Lynn's example and remove the DC component from the modulation and take a look at the result.  Granted this is considered double-sideband suppressed carrier, but it is still AM modulation.

If you know enough, then you can use it.  But the original poster mentioned some kind of noise analysis, so I don't know if this is one of those cases.

[johnsold]

Randall is correct. I assumed, without being explicit about it, that since envelope detection was requested the modulation type was one where it would be suitable, namely amplitude modulation with full carrier and modulation index less than 1.

Lynn

[Eng P&D]

I read about hilbert envelope based on Im numbers, and make this one.

could some tell me if I'm taking right way?

thanks

[rpursley8]

Replace the absolute value function with a complex to polar function and output of magnitude (r) and you will have it.  When you combine a signal with its Hilbert pair signal in the complex form, it is called an analytic signal and will provide the envelope (magnitude) of the input signal.

Message Edited by rpursley8 on 02-06-2006 01:31 PM

[tbob]

Seems as if there is a little misunderstanding of questions and answers.   I was just pointing out that a couple of your statements were wrong:  Using absolute value does not double the modulating signal, and the detector stage does not need to know what the carrier frequency is.  I guess this is getting off track because it does not pertain to the original question.  Of course Lynn's software is only good for an AM signal.  I believe he was just trying to give an example of how an envelope can be extracted from a modulated carrier using AM as an example.

I am missing something in another statement you said.  Why must the modulating signal be DC shifted?  In Lynn's vi, if you remove the DC shift by changing the 1 to a 0 in the addition function, you still get an output that is similar to the modulating signal, just lower in amplitude.  Does the DC shift have something to do with the mixing of the carrier and modulating signal?  Please clarify.

[rpursley8]

The demodulated signal is double the frequency too when you remove the DC offset.

Message Edited by rpursley8 on 02-06-2006 01:54 PM

[johnsold]

Try using an assymetric modulating function and you will see what we are talking about. Also, if you look at the carrier at the zero crossing of the modulating waveform you will see a phase reversal.

Lynn

[tbob]

Now I get it.  Its a little hard to see, but I can see the phase reversal at the zero crossing.  Thanx for the explanation.  In practical experience, I have never seen the modulating signal go negative so I didn't know of that effect.