How to multiply the measured signal with exp(-t/T) and integrate it?

[Chunrok]

For my optimum filtering, I should multiply the measured signal S(t) with exp(-t/T) and integrate these from 0 to tmax. Here, T is the time constant tau.
Does anyone know how to do it in LabVIEW?

[unknown]

I build a vi that implements your formula. More comments are in the vi.
If you have problems please send me a message.

[altenbach]

I guess your problem description is not entirely clear.

So we are starting out with an array S(t). After exponential multiplication and integration, we are left with a single value, according to your description, so it's more "data reduction" than filtering.

Are you sure that's what you want? What exactly should the result be, I can think of a few possible alternative interpretations, e.g.:

(1) You are looking for the scalar value of the integral.
(2) You are looking for a new signal where each point is the integral up to that point.
(3) You are looking for a point-by-point type implementation, where you apply this function to a sliding window, replacing the signal with the filtered values as you go.
(4) etc.

In any case, all this is simple
math and the LabVIEW diagram will probably fit the size of a postage stamp.

Please give a bit more detail and I'll wire something up.

[Chunrok]

Wow, you are cool!

At first, I am sorry if I wasn't clear enough. It was probably because I didn't understand my problem and LabVIEW exactly.

Anyway, I think what I want looks like (3). However, I would also like to know other possibilities, at least (1). Because I will need it eventually to calculate the energy of my each pulse signal.

I appreciate your help. I really mean it. Thanks much.


Chunrok Bae

[Chunrok]

Thanks much for your valuable answer.
It gave me a good idea. However, I need more or less like a point-by-point type implementation to replace my measured pulse signal with its filtered value.

Chunrok

[Chunrok]

I don't know how I can reach you or send you a message in this NI site.
Could you please help me a little more on my problem?

Chunrok

[Chunrok]

Hi, you looked like being ready to help me about my problem.
However, you haven't posted any answer yet.
Could you please help me a little more on my LabVIEW programming problem?

Chunrok

[unknown]

To "Chunrok":
You can send me a personal mail at nistvanrobert@yahoo.com with your problem. If you want to send me a vi please save as compatible to Labview 6.

[altenbach]

Sorry, this slipped off my horizon ...Thanks for the reminder.

OK, while I don't understand the reasoning behind the algorithm, I have implemented Case (1) and Case (3) with a simple demo using static data. It's just simple math, nothing fancy.

(I haven't fully tested these, so there might be bugs. They are in LabVIEW 7.1, do you need a lower version?)

[Chunrok]

Thanks much!
Unfortunately, I have LabVIEW 7 Express so that I can't open the demo you attached.
Could you please inform me in LabVIEW 7?

Chunrok