An analog signal refers to information represented by continuously varying physical quantities—such as time, amplitude, frequency, or phase. Analog signals are ubiquitous in nature, appearing in everyday phenomena like daily temperature fluctuations, the speed of a car while driving, or the voltage amplitude at a specific node in an electrical circuit. 
Digital signals are artificially abstracted, discontinuous signals that can typically be derived from analog signals. The values of a digital signal are discrete and finite in number. 
Simulating the digitization of an analog signal involves converting the analog signal into a discrete sequence that can be represented by a finite set of numerical values. But exactly how does this process work? The image below provides the answer.
 

The process of digitizing analog audio signals 
Converting an analog audio signal to a digital audio signal: An analog audio signal is one that is continuous in both time and amplitude, and its digitization process is described below. 
1. Sampling: Digitizing a signal along the time axis. Specifically, this involves extracting the values of an analog signal at fixed time intervals, effectively transforming a continuously varying information waveform into a discrete signal with a finite number of samples over time. 
2. Quantization: Digitizing a signal along the amplitude axis. In other words, approximating the originally continuously varying amplitude values using a finite set of amplitude levels, thereby converting the continuous amplitudes of the analog signal into a limited number of discrete values spaced at regular intervals. 
3. Encoding: Represent the quantized value (in decimal) of each sample using binary numbers. 
Simulating the digitalization process of video signals 
Converting an analog video signal into a digital video signal: An analog video signal is continuous both spatially and in terms of grayscale, and its digitization process differs from that of audio signals. 
1. Sampling: Digitizing a signal in space. In other words, the image is represented by grayscale values at selected points across the spatial domain—these points are called sample points (or pixels). During sampling, the total number of pixels along both the vertical and horizontal axes of the image directly determines the quality of the digital image. Unlike audio signals, where amplitude values are sampled in the time domain, video signals involve sampling the function that defines the image itself. After sampling, the image is divided into discrete spatial pixels, though the grayscale values within each pixel remain continuous. 
2. Quantization: Digitizing a signal on a grayscale scale by converting pixel intensities into discrete integer values. 
The number of distinct gray levels in a digital image is referred to as the grayscale level. For instance, in the RGB888 color model (R: red, G: green, B: blue), each color component can take on 256 possible values, resulting in a grayscale level of 256. Since 256 equals \(2^8\), each color can be represented using an 8-bit binary number, effectively quantizing the image information into binary form. From a visual perspective, images with grayscale levels quantized to 6 bits or more already produce satisfactory results, as the human eye typically can distinguish up to about 100 gray levels at most.