We are looking at FMCW radar ADC signals and converting from the time domain to the frequency domain in order build a range doppler map but want to have the best signal to noise ratio possible.

In order to reduce spectral leakage, we are using the Hann function. It just seems to be accepted that we use the Hann function.

Are there any other functions that we should be considering?

  • 2
    $\begingroup$ no, it's not "generally accepted that we use the Hann function". Choice of window depends on what you want to minimize (e.g. power in strongest sidelobe OR cumulative power out-of-band OR attenuation at $f\rightarrow \infty$ OR …). $\endgroup$ Jul 6, 2017 at 13:21
  • $\begingroup$ @MarcusMüller thanks. We're building a doppler range map. I've edited the question to reflect. $\endgroup$
    – SeanJ
    Jul 6, 2017 at 13:29
  • $\begingroup$ A friend of mine wrote a paper on Range/velocity diagrams and windowing, can't think of the title. Look for papers by Stefan Wunsch! $\endgroup$ Jul 6, 2017 at 13:30
  • $\begingroup$ Improved detector for continuous wave radar waveforms exploiting phase information - sadly don't have an IEEE membership ($33 for paper)! Is it the right article? $\endgroup$
    – SeanJ
    Jul 6, 2017 at 13:47
  • 1
    $\begingroup$ @SeanJ A simple Google search of the author on that IEEE paper shows that he makes a preprint available: cel.kit.edu/download/Radarcon_2015_final_fonts_embedded.pdf $\endgroup$
    – Peter K.
    Jul 6, 2017 at 17:02

1 Answer 1


If you want to have the best SNR possible, you put a null on the signal that is leaking it's energy. In order to do that, you either have to know the location of the interferer or use an adaptive filtering scheme.

If you want to stick with a fixed window, the Taylor window isn't a bad choice, It's like a Chebyshev and rolls off further out.

A. Villeneuve, "Taylor patterns for discrete arrays," in IEEE Transactions on Antennas and Propagation, vol. 32, no. 10, pp. 1089-1093, October 1984.


Your Answer

By clicking “Post Your Answer”, you agree to our terms of service and acknowledge you have read our privacy policy.

Not the answer you're looking for? Browse other questions tagged or ask your own question.