The real time sampling is converting the analog signal to digital bits at a fixed rate. There is no relationship between the sampling frequency and the frequency of the signal being converted. The time between the digital samples is exactly the same and equal to the sampling time. The digital representation of the original analog signal will then have missing sections between the consecutive samples. When the sampling frequency is much higher than the analog signal frequency (>100,) the digital construction of the signal is an accurate model of the input. However, as the input frequency increases, the constructed digital signal shows signs of discrepancy due to the missing information between the samples. For example, when the sampling frequency is exactly equal to the analog signal frequency, there is only one point of digital information available to construct the signal from. This will result in a DC line. Clearly, the digital model at higher input frequencies is far from the original signal.
The effective sampling rate is referred to a sampling technique for periodical signals. In this technique, each time the analog input is sampled, the sampling signal is delayed by a predetermined amount (ΔT) with respect to a fixed point of the input. Therefore, the time resolution between the consecutive samples is effectively ΔT, yielding a sampling rate of 1/ ΔT that is independent of the frequency of the sampling signal. The tolerance of ΔT over the consecutive samples determines the accuracy of the reconstructed model.