• State Resolved
  • Locked Locked
  • Replies 5 replies
  • Answers 1 answer
  • Subscribers 29 subscribers
  • Views 291 views
  • Users 0 members are here
Support feedback
Options
Options
Related

This thread has been locked.

If you have a related question, please click the "Ask a related question" button in the top right corner. The newly created question will be automatically linked to this question.

MMWCAS-RF-EVM: Purpose of the range scale factor coefficients

Dayalan Nair
Prodigy 220 points
Part Number: MMWCAS-RF-EVM

Hi, I am not sure what the purpose of the scaleFactor in the module_param.m script is. Please assist. 

scaleFactor         = [0.0625, 0.03125, 0.015625, 0.0078125, 0.00390625, 0.001953125, 0.0009765625, 0.00048828125]*4;

.

.

.

rangeProcCascade_scaleFactorRange = 0; 
rangeProcCascade_FFTOutScaleOn = 0; %1: apply scaleFactorRange; 0: scaling factor not applied

  • 0
    TI__Guru*** 148804 points

    Hi,

    Unfortunately the developer of this matlab code is no longer available for support.

    That being said, it looks like the scaleFactor array is as follows

    (1/16; 1/32; 1/64...)

    These are probably used for scaling the range FFT to avoid overflow

    thank you

    Cesar

  • 0 in reply to Cesar
    Prodigy 220 points

    Hi, 

    Could you please elaborate on how overflow is avoided using this method? There does not seem to be any code that checks rangeProcCascade_FFTOutScaleOn. Why are there 8 values and why are they multiplied by 4? I understand it is not your code, would like your opinion.

    Thanks

  • +1 in reply to Dayalan Nair
    TI__Guru*** 148804 points

    Hi,

    The unit building block to compute FFT is the FFT butterfly. In this operation, due to the Multiply Accumulate operation overflow can occur.

    Usually to avoid overflow scaling by 2 is performed after each radix-2 stage.

    Thank you

    Cesar

  • 0 in reply to Cesar
    Prodigy 220 points

    Hi,  

    Yes that makes sense. It must be a radix-2 DIT algorithm which is used in the MATLAB fft() function. I am still not sure why these values are multiplied by 4, would be nice to know but not critical at the moment.

    Thanks

  • +1 in reply to Dayalan Nair
    TI__Guru*** 148804 points

    Thank you

    Multiplying by 4 is also probably a scaling

    Unfortunately the developer of this code is no longer available for support

    Thank you

    Cesar