Signal discretization?

Question

Signal processing

There is a signal, do I understand correctly that if the signal is limited, then the sampling step is chosen through the Kotelnikov theorem, and if it is not limited. then through the Zheleznov criterion?
Is this exactly what is called "this is how sampling steps are chosen in practice"?
What is the sampling step?

Answer 1

In practice, you choose the sample rate. An unlimited signal can only be in the imagination, IMHO.
BH is selected according to the following criteria:
1. Based on the maximum frequency of the used frequency range. The sampling frequency is obtained by the Kotelnikov/Nyquist theorem. Those. simply multiply by 2 the maximum frequency of your range. For high frequencies there are some nuances. The resulting black hole is the lower bound for a real black hole.
2. Characteristics of the input signal. High-frequency noise may be present in the input signal, if you do not filter them with an analog filter, then this must be taken into account when choosing a frequency response. It is also necessary to take into account the frequency response of the analog filter, because there is nothing perfect in life.
3. The capabilities of the equipment used. ADCs can produce far from arbitrary black holes. ADC capabilities depend on many details of a particular model. Usually, when you initialize the ADC, you set the desired frequency, and the ADC works on the one on which it can and which is most suitable for your Wishlist. So after initialization, you always need to poll the ADC on what kind of black hole it works.