International Electronics and Technology Forum
May 21, 2012, 01:48:02 PM *
Welcome, Guest. Please login or register.
Did you miss your activation email?

Login with username, password and session length
News:
 
   Home   Help Search Login Register  
Pages: [1]
  Print  
Author Topic: FFT on I/Q-data Vs just I-data?  (Read 433 times)
Rohit_K
Newbie
*
Posts: 4


« on: July 16, 2011, 07:37:51 AM »

I'm trying to obtain a frequency analysis of a signal in the 40Khz-50Khz band. To do this I first mix it with a 40Khz square wave LO & 4th order elliptic low-pass filter. Then feed the output to a single ended ADC sampling at 20Ksps on a microcontroller programmed with a DFT algorithm.

HOWEVER, i never seem to get totally consistent results on the spectrum. My question is this - would it make a difference if I mixed/ downconverted it to get I/Q signals & then do a complex DFT ? Currently I am doing the equivalent of using only the "I" signal. What would be the difference in terms of frequency analysis if any ?

thanks!
The focus is on what would be the difference between DFT output on just the I-data vs DFT output of I /Q- data.
Even so, how would the phase shift of LPF affect DFT analysis output ?
Logged
veeyesvee
Newbie
*
Posts: 125


« Reply #1 on: July 16, 2011, 08:21:04 AM »

Did you look into the phase shifts caused by the elliptic filter?? Perhaps you should use a bessel filter of as lowest order as necessary. The phase shift does affect both I and Q signals equally. So if you do I and Q analysis, you might get expected results if you also look at phase shift created by the post conversion filter.
Logged
Pages: [1]
  Print  
 
Jump to:  

Powered by MySQL Powered by PHP Powered by SMF 1.1.13 | SMF © 2006-2011, Simple Machines LLC | Privacy Policy Valid XHTML 1.0! Valid CSS!