Phase uncertainties and signal sidelobes

[Jasondspnoob]

Hi all, I am seeking a correlation between phase and sidelobes (formulas would be appreciated). Would phase uncertainties cause worse sidelobes? Are they related? If they are, one would think uncertainties are random (which can make sidelobes better or worse).

Thank you for your help! :smile:

Jason

 

[permute]

Look into jitter analysis of ADC's, Phase modulation, and DDS systems. Each can be helpful.

the uncertainty generates sidelobes for the same reason phase modulation does. The equivalent index is low, so you probably will only see one sidelobe. The effects become more pronounced for high-speed ADC's sampling signals above Fs/2 (bandpass sampling). The DDS's also have similar phase-quantization issues. In such cases, spurs are generates reducing SFDR. in that case, randomizing the phase slightly tends to have the effect of spreading the phase error out over the entire spectrum.

things like power supplies can affect the phase of a single-ended clocking system, resulting in a slightly modulated clock and thus small sidebands at the power supply switching frequency.

in the DDS case, the algorithm has quantization, which can either be correlated (resulting in spurs), or uncorrelated (resulting in noise).

I'm not sure of your application.

 

[Jasondspnoob]

Thanks for the guidance. My application is to use labview to transform frequency response (s21 measurements from network analyzer) into the time domain. A pulse is sent to a narrowband filter. somehow it seems like if the uncertainty in phase would bring the sidelobes higher, it would also bring the mainlobe higher. Does it make sense?