Digital I/O

cancel
Showing results for 
Search instead for 
Did you mean: 

FFT analysis on FPGA

Hi ,

 

I want to analyse FFT Spectrum on FPGA. First of all, i would like to calculate the amplitude ( imag^2+real^2 = amplitude) of the FFT calculated values.(made up here)  then found to be the biggest value and corresponding index. How can I do this ?

 

regards,

Mehmet

0 Kudos
Message 1 of 6
(4,824 Views)

maybe you could use XILINX IP CORE

0 Kudos
Message 2 of 6
(4,810 Views)

Hi Mehmet,

 

in order to provide you with some help it would be useful to get some more details regarding your case.

 

1) What kind of hardware do you use? NI Hardware? CompactRIO, FlexRIO, etc. ..? Modules?

2) What kind of software do you use? LabVIEW? FPGA-module? Please also provide version information.

 

3) What kind of signal is your input? Is it already available in digital form or still analog?

4) Which measures to solve the problem did you take up to now? Did you use the "Auto Pwr Spectrum" function from "Signal Processing" in LabVIEW to calculate your spectrum?

 

Let's try to solve your problem together. 🙂

 

Best regards,

 

Benjamin

 

0 Kudos
Message 3 of 6
(4,801 Views)

Hi Mehmet,

I did a internet search on your problem and found some information that might be of interest for you.

You can find a thread discussing the same problem here: Peak Detection Algorithm for FPGA LabView

More general information on "peak detection" using LabVIEW can be found here: Peak Detection Using LabVIEW and Measurement Studio


Here you can find a VI from IPNet that could exactly solve your problem: Frame Max/Min

 

Let me know if this helped.

 

Thanks,

 

Benjamin

0 Kudos
Message 4 of 6
(4,788 Views)

Hi Benjamin ,

 

I use 7853R with Labview 2010 and FPGA Module. It is a analog signal.

I want to solve the problem from amplitude spectrum. I send you a screenshot from my FPGA-Programm.

 

regards,

Mehmet

Download All
0 Kudos
Message 5 of 6
(4,767 Views)

Hi,

 

Why are you using this target-scopd FIFO's? The way you use them doesn't make sense.

Use the Scale by Power of 2 instead of the Quotient and Remainder function since it takes less space and only 1 tick instead of 32 ticks if you divide by a number which is a power of 2.

 

I would recommend you to have a look at one of the FFT examples shipped with LabVIEW.

You will find them under <labview-folder>\examples\R Series\FPGA Fundamentals\Analysis and Control\Fft

 

 


Christian

0 Kudos
Message 6 of 6
(4,759 Views)