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.

  • Resolved

MSP430FR5994: TIDM-FILTERING-SIGNALPROCESSING "512-point FFT with LEA"

Intellectual 865 points

Replies: 3

Views: 686

Part Number: MSP430FR5994

Hi,
 
We are testing FFT with LAE with the sample codes from the TI design
"TIDM-FILTERING-SIGNALPROCESSING".
 
This sample code is designed for "256-point complex FFT" but we need 512-point.
Is there any way to change this sample code for 512-point FFT calculation?
 
Tried to change the "#define VECTOR_SIZE" to 512, but it gives error saying 
LEARAM overflow.Please let us know what setting is necessary for 512-point FFT.
 
Best Regards,
Tadaaki Matsumoto
  • Hi Tadaaki,

    An error is thrown when you change the vector size to 512 because it can no longer properly align all the different variables in LEA RAM. This example can be configured to perform either an FFT, FIR, or IIR on the input data so all these variables must be able to fit into LEA RAM with the proper alignment.

    If your end goal is to just test out a 512 point FFT to see if it works, I would suggest looking into the example provided in DSPLIB and changing the sample size to 512:
    software-dl.ti.com/.../transform_ex1_fft_fixed_q15_8c-example.html

    Otherwise, you'll need to modify the example to remove the FIR and/or the IIR capability, freeing up space in LEA RAM.

    Best regards,
    Caleb Overbay
  • In reply to Caleb Overbay:

    Hi Caleb-san,
    Thank you very much for your quick response.
    Our customer would like to modify the example to remove the FIR and/or the IIR capability, freeing up space in LEA RAM.
    And they have an additional question on this.
    Could you please let us know How to change "const _q15 hammingWindow[VECTOR_SIZE]" in FFT.c from for 256 points to 512 points?
    Best Regards,
    Tadaaki Matsumoto
  • In reply to tadaaki matsumoto40:

    Hi Tadaaki, 

    I apologize for the late reply. The hammingWindow in the TI design is hard-coded so it's not very straightforward to change. Instead I recommend looking at the example linked below which can construct a hamming window of variable length:

    http://software-dl.ti.com/msp430/msp430_public_sw/mcu/msp430/DSPLib/latest/exports/html/transform_ex1_fft_fixed_q15_8c-example.html

    Best regards, 
    Caleb Overbay

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.