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.

Need help passing FFT arguments/vector



Hello,

i am a student and new to DSP programming and need help. My exercise is to implement a fft on the C5509 with the TMS320C55x DSP Library. For introduction i am using a matlab-generated sinusoid, which i am loading into memory using Code Composer Studio. The fft is doing something with this vector, but the output doesn't make any sense. I guess there is something wrong with my data vector or with my arguments. I am also not sure if i got all the information in the TMS320C55x DSP Library Programmer’s Reference.

My sinusoid has 256 real-value samples, so my re-im interleaved matlab-file looks like this:

0x3333
0x0000
0x24C4
0x0000
0xFD94
0x0000
0xE238
0x0000
.....

(imaginary values = 0).

When calling the cfft (noscale), i am passing n = 256 (vector length). I am expecting 2 different frequencies, but the output is something similar to my input.

Is there any obvious error in my number format, vector, arguments?

Thanks a lot!

Regards,
Florian

  • HI Florian,

    I'll give your numbers a try to see if I can reproduce what you have described. Let me get back to you.

    Thanks,
    Mark

  • Hey,

    finially my sourcecode ist working!

    The problem was very simple, but because i am a totally beginner and because it is not mentioned in the DSP library reference, i just did not notice it:
    CPL Mode was not enabled, and so the relative adressing within the cfft-routine did not work. The arguments were stored in IER0 and IFR0 instead on stack.

    Thanks for your support!