Dear UT Experts,
I am currently trying to get a better picture of how UT NDT Inspectors determine the noise level. This is why I have started this survey and I would like to ask you to provide YOUR personal view on this subject.
My question: Looking at the signal in the attached figure - which level would you report as the noise level? 0.3; 0.4; 0.5; 0.6; 0.7; 0.8; 0.9; 1.0? Or something in-between?
Thank you ...
It's common to talk about the noise power, which would be the RMS of that curve. That will tend to bias it towards the higher linear levels, so you can figure something closer to the 0.6 level instead of the 0.45 level or so where the linear average would appear to be. This directly relates to the accuracy of a power measurement of a signal (averaged over time) that will be the sum of the signal power and the noise power.
Now, the other side of this equation is talking about signal-to-noise ratio (SNR) and it's impact on a voltage measurement. In communication, they typically talk about EVM (error vector magnitude) which is a voltage related (linear) error term caused by the vector sum of the signal + noise. The error in a voltage measurement (in dB) follows a 20 log relationship rather than the 10 log relationship for power.
If not particularly specified we prefer RMS only and peak should be preferred as the highest in the window as maximum possible noise. So the peak can be 1V. For S/N just consider same levels for signal and noise
- Badges
- Science method