UPGRADE YOUR BROWSER

We have detected your current browser version is not the latest one. Xilinx.com uses the latest web technologies to bring you the best online experience possible. Please upgrade to a Xilinx.com supported browser:Chrome, Firefox, Internet Explorer 11, Safari. Thank you!

cancel
Showing results for 
Search instead for 
Did you mean: 
Highlighted
Visitor jojo89
Visitor
500 Views
Registered: ‎12-21-2017

Incorrect m_axi_data_tlast signal from fft-Block

Hi guys,

 

I'm currently doing an design in System generator, but I've got some problems with the Xilinx Fast-Fourier-Transfrom 9.0.

 

My aim:

I want to do a fft on different time frames of my input signal.

I don't want to change my transformation length, though.

This is why I tried to solve this issue by downsampling.

 

As the downsampling block doesn't support variable sample times, I used another approach:

I provide an appropiate pulse at the "s_axis_data_tvalid" Port.

When I want to downsample by 2, I provide a pulse that equally toggles between 0 and 1.

 

My problem:

The output "m_axis_data_tlast"  is quite wrong.

The spectrum doesn't seem right, too.

 

One concrete example:

I use a sine wave (period: 16) as input signal:

sineWave.png

This shows my sine wave (yellow) and the signal after passing the xilinx input port(blue)

 

My Transformation length is 4096, I want to downsample the signal by 2.

pulse.png

This is my pulse signal (yellow). The red one is the enable signal: this is the time when everything should start.

 

LastInput.png

This is my input signal for s_axis_data_tlast. Distance between start and the peak is 8190, which is correct.

 

outputLast.png

This is the output m_axix_Data_tlast signal. As you see, the distance between two peaks seems right.

 

outputLast2.png

Anyway, as you look closer, you see, that there are two peaks. 

The distance is 33 samples, as the next screenshot shows.

 

outputLast3.png

 

Attached you find a minimal example of my problem.

Thanks in advance for helping me solving this issue.

If you need any further information please let me know!

 

Tags (2)
0 Kudos