How to separate and group string?

[yatang]

Hi Jeff, please have a look on the attachment, I build array for FFT, but there's still no result displayed on the graph.

Thank you.

[JÞB]

Well you have some configuration settings incorrect on that express vi.  Of course, if you hadn't tried to use Dynamic Data Type despite the warnings prevelant in the forums, you would have found it a lot easier.  

That express vi could probably handle the RMS averaging too if you convert it to a normal vi and wire in the FFT size.  Not that this should matter!

You need to create a rolling buffer for your array.  You started out well by initializing the array but then appended to it rather than use the replace array subset function.  Replace array (if used properly) will not cause the array size to constantly grow.  In this example the latest data is always placed in element 0

And just to channel my inner Altenbach, This is a perfect example of when to use CDB data . It even works with express vis, graphs, advanced math functions and charts.

See the modified code for comments.  (You will need to relink some missing type defs)

[yatang]

Jeff, since my labview is 2012, can you change the VI to 2012 version?

Thank you very much.

[JÞB]

---

@yatang wrote:

Jeff, since my labview is 2012, can you change the VI to 2012 version?

Thank you very much.

---

Sorry about that-  I should have caught the originating version from the VI properties.  (Often I do but, today I was also distracted by work.)

Here is the vi back saved to LabVIEW 2012.  Again, it is going to need some re-linking to the missing type defs and I have no means to generate sample data for quality assurance but, it should be instructive. on how to greatly simplfy the code by using functions you may not have looked deeply at yet.

[JÞB]

well

I have often bragged about my ini file option to show coersion dots in a grotesque color..........

perhaps those dots should be larger  ....

OP, that will not work as I provided... the CDBs are getting coerced.......to DBLs

Spoiler

So much for "How would Christian do that?"  DAMN.. none of those key functions have a CDB override?

As an asside-  I used to hold a job title of "Jr radar system engineer" at a company that designed weather radars

Spoiler

Don't belive me ---- Really, I spelled"weather" correctly, in context, on the first try.

[yatang]

LOL, thank you very much! I wil deeply study your code!

[yatang]

Jeff! You are really amazing!! I learnt a lot from your code! 

And I still want to ask you one more question, if I want to use "waveform graph" to display the I/Q signal, how can I modify the code? I still need to build a array?

[yatang]

And please have a look on this picture, I separately display I and Q signal on the chart, you will find that Q signal is more like a square wave, I don't know why T_T, the labview code for I/Q signal is absolutely the same.

And there's seems no results for FFT(I changed FFT to linear scale).

And as I mentioned above, if I hope to use "waveform graph" to display I/Q signal, should I also build a arrary?

[JÞB]

Just a small edit to take the IQ CDB array through "Complex to Re Im" (On Programming>>Mathematics>>Complex sub palatte) then bundle Re and Im arrays into X and Y for your XY Graph

There is also a Compass plot that can has a polymorphic instance of 1D Complex. It depends on how you want to see it

Edit: I just saw the latest post

With those I and Q signals I would be very surprised if you could find a frequency in the power spectrum (but it looks like you got a blip on the radar.

I'm thinking something is very off in your dT and double check me on this.

at 256 samples (Your FFT Size default) with a sample rate of 1/20 seconds your rolling buffer covers 12.8 seconds of acquisition.

at 12.2uSec/Radar mile thats almost 1.05 MILLION miles! or 4 times the distance to the moon.  That's a heck of a radar buddy!

[yatang]

I think I solve the FFT problem! 

But do you know why the Q signal in the picture is like square wave?