DSPRelated.com

(@Justinm)


Re: DTFT optimization question

Reply posted 3 years ago (06/03/2021)
Looking more at a radix FFT implementations I don't know if the sub-sequence coefficients are necessary intermediate results.  Some of the implementations use...

Re: DTFT optimization question

Reply posted 3 years ago (06/03/2021)
I think I can describe the desired FT coefficients more precisely. I am given the 8192 sample time series (sampled at 1 GHz) and 10 frequencies (These frequencies...

Re: DTFT optimization question

Reply posted 3 years ago (06/03/2021)
thanks for the reply.I am partitioning the sequence into 4 sub-sequences 0-2047, 2048-4095, 4096-6143,6144-8191.  I have real input and compute complex coefficients.I...

DTFT optimization question

New thread started 3 years ago
Hello Everyone,I am using cuda on a Jetson Xavier AGX to do some signal processing. I am processing a time series which is 16 bit samples taken at 1 GHz. Each...

Re: hardware FFT coefficients .vs. np.numpy.fft.rfft

Reply posted 3 years ago (12/19/2020)
Thanks for the continued thought about my problem.  We came to the same conclusion.The FPGA engineer determined the trigger starting the FFT was one clock cycle...

Re: hardware FFT coefficients .vs. np.numpy.fft.rfft

Reply posted 3 years ago (12/15/2020)
You got it. :)  We set up a testbench for the FPGA and fed in a set of test data.  We ran the FFT on the FPGA and compared the results with the FFT I was generating...

Re: hardware FFT coefficients .vs. np.numpy.fft.rfft

Reply posted 3 years ago (12/14/2020)
We think we have the same time series for both the FPGA FFT and the NP FFT.  The FPGA sends me the coefficients it calculates and the time series.  I thought that...

Re: hardware FFT coefficients .vs. np.numpy.fft.rfft

Reply posted 3 years ago (12/14/2020)
I asked our DSP guy about this and he thinks it is worth checking.  He suggested putting in a test signal and looking for the phase walk.  Do you have any ideas...

Re: hardware FFT coefficients .vs. np.numpy.fft.rfft

Reply posted 3 years ago (12/14/2020)
I looked at the spectrum of both FFTs, (the FPGA and the numpy.fft.rfft) and the magnitudes look the same.  When you say order I think that means low frequency...
We have a FPGA system which takes input from an ADC and calculates a FFT. The system identifies the location of our frequencies of interest and sends the coefficients...

Use this form to contact Justinm

Before you can contact a member of the *Related Sites:

  • You must be logged in (register here)
  • You must confirm you email address