Re: Converting FFT to inverse-FFT

On 12/15/2009 2:50 PM, HardySpicer wrote:
On Dec 16, 3:22 am, Vladimir Vassilevsky<nos...@xxxxxxxxxxx> wrote:
Philip Pemberton wrote:

How would I go about converting this from an FFT to an inverse-FFT?
Scale by factor 1/N and use -sin(x) instead of sin(x).

Sorry if this seems like a stupid question, it just seems a little too
easy...!
Stupid question indeed.

Vladimir Vassilevsky
DSP and Mixed Signal Design Consultanthttp://www.abvolt.com

In fact the ordinary FFT should have a 1/N, not the inverse. Makes
more sense. For example for DC you need to divide by N ie the average.


Hardy

It's just convention. In the old days we put 1/sqrt(N) in front of each. There are some advantages to that, and some disadvantages. As long as one keeps track, it comes out in the wash.

--
Eric Jacobsen
Minister of Algorithms
Abineau Communications
http://www.abineau.com
.

Relevant Pages

  • Re: Favorite reverse bias protection for battery circuits
    ... Vladimir Vassilevsky ... DSP and Mixed Signal Design Consultant ...
    (sci.electronics.design)
  • Re: divide-by-N
    ... Output duty cycle doesn't matter. ... Vladimir Vassilevsky ... DSP and Mixed Signal Design Consultant ...
    (sci.electronics.design)
  • Re: AFT (Arithmetic Fourier Transform)
    ... Vladimir Vassilevsky ... DSP and Mixed Signal Design Consultanthttp://www.abvolt.com ... Requires non-uniformly sampled input so you need ...
    (comp.dsp)
  • Re: DSP implementation on an FPGA: Best HDL?
    ... Verilog would be an appropriate choice for this DSP design. ... DSP and Mixed Signal Design Consultanthttp://www.abvolt.com ... What preliminary work do you think gave him to know that his extended Kalman filter needs 17 states, ...
    (comp.dsp)