In digital signaling, the set values are discrete (i.e. 0's or 1's); whereas in analog, values are continuous over a given range.
Now, let's take an example to show the difference between analog and digital signaling. If we take an analog voltage range between 0 to +15V where the analog signal represents an infinite number of values within the range then the 4-bit representation of this signal would have 16 discrete values starting from 0000 to 1111 (binary code). If we increase the number of bits in the digital code to 8 or 16 bits, then the number of discrete values would also increase to 256 and 65536 respectively. Thus, the greater the number of bits in the digital code, the greater the accuracy of the analog signal representation. This is illustrated in Figure 1 where the analog signal is shown as a smooth curve that is within the range of 0 V and +15 V and where a 4-bit code represents a discrete number of points on the curve.
In Figure 1 the analog voltage is sampled and measure at 35 equal intervals. The output voltage is represented by a 4-bit binary code with 16 discrete values. Hence, a series of digital representation of the analog input voltage is represented along the analog curve. This illustrates how an Analog-to-Digital Conversion works. In addition, an approximation of the analog signal can be taken from Figure 1 and reconstructed to form the 4-bit digital code of discrete values. It can be seen that there is a margin of error when the conversion occurs due to the sample size (i.e. the number of bits used to represent the analog signal) where only certain values are used.
Now if we look at Digital-to-Analog conversion, where the digital values are plotted on the graph as shown in Figure 2, the reconstruction of the analog signal can be seen. Note that this is only an approximation because not all the values of the points are available. Thus it can be concluded that increasing the sample size or increasing the number of bits will increase the accuracy of the analog signal.
