Signal Processing Stack Exchange is a question and answer site for practitioners of the art and science of signal, image and video processing. It only takes a minute to sign up.
Sign up to join this community
What do the following statements on pages 13 and 15 in this linked document mean?
DNL, a static specification, relates to SNR, a dynamic specification.
INL is a static specification and relates to THD (a dynamic specification).
I don't quite understand what kind of connection it is. What is the relationship between practice and theory? Your details would be appreciated.
DNL adds extra noise, in addition to the quantisation noise. If the DNL of an ADC is $\pm 0.5$ LSB and is uniformly distributed between these values, then the DNL will add extra $\frac{LSB^{2}}{12}$ of extra noise over the quantisation noise. In this particular case, the total noise floor will be $3$dB higher than the quantisation noise. If the DNL is higher, then the noise floor is raised higher and thus the total SNR is decreased. In general the DNL will have a gaussian distribution.
INL is the transfer function deviation from the straight line. So it adds distortion components to the ADC output. Hence the impact on THD. The THD is roughly calculated as $20 \log_{10} \left( \frac{INL}{2^{N-1}} \right)$ where INL is expressed in LSBs.
For an 16 bit ADC with $1$ LSB of INL, the best case THD (at low frequency) will be about -90dB.
