![[4-bit-linear-PCM.svg]] Example of a signed 4-bit integer quantizer ([Wikimedia](https://commons.wikimedia.org/wiki/File:4-bit-linear-PCM.svg)) In analog-to-digital conversion, a [[continuous-time signal]] is converted into a digital signal by [[sampling]] the analog signal at regular intervals and then quantizing the sampled values. Amplitude quantization is the second step in this process, where the continuous-time signal's amplitude values are rounded or truncated to a specific number of bits. The number of bits used to represent the amplitude values determines the amplitude resolution of the digital signal. For example, if a 16-bit quantizer is used, the amplitude values can only take on $2^{16}$ (or 65536) discrete levels. If the amplitude is represented as a signed integer, the range of discrete amplitudes will be in the range $\{-32767, \ldots, 32768\}$. The process of regularly sampling and linearly quantizing the analog signal is called linear PCM (Pulse Code Modulation). It is common to quantize amplitudes in 16, 24 or 32 bits.