PCM
Definition
PCM (Pulse Code Modulation) is a digital communication technique in which an analog message signal is converted into a digital bit stream by sampling the signal at regular intervals, quantizing the sampled values into discrete amplitude levels, and then encoding those levels into binary form.
In simple terms, PCM is the most fundamental method used to represent an analog signal digitally. It is widely used in telephony, audio recording, compact discs, digital communication systems, and modern data transmission because it provides high noise immunity and reliable signal reproduction.
Main Content
1. Sampling
Meaning of sampling
- Sampling is the process of measuring the amplitude of an analog signal at equal time intervals. The continuous waveform is converted into a sequence of discrete-time values.
Why sampling is needed
- Digital systems cannot directly handle continuously varying analog signals, so the signal must first be broken into small time-based snapshots.
Sampling frequency
- The number of samples taken per second is called the sampling frequency or sampling rate, measured in hertz (Hz).
Nyquist theorem
- To reconstruct the original signal accurately, the sampling frequency must be at least twice the highest frequency component present in the signal. This is called the Nyquist criterion.
Example
- For audio with a highest frequency of 4 kHz, the minimum sampling frequency should be 8 kHz.
Effect of low sampling
- If the sampling rate is too low, aliasing occurs, where higher-frequency components appear as false lower-frequency components in the reconstructed signal.
Sampling illustration
Analog signal: ~~~~~~~~ continuous waveform ~~~~~~~~
Sampling points: | | | | | |
Discrete samples: x x x x x x
2. Quantization
Meaning of quantization
- Quantization is the process of assigning each sampled amplitude to the nearest fixed discrete level.
Why quantization is required
- A sample may have any value within a range, but digital systems require finite levels, so each sample is approximated to the nearest permitted level.
Quantization levels
- If the number of levels is , then each sample is mapped to one of these values.
Quantization error
- The difference between the original sampled value and the quantized value is called quantization error or quantization noise.
Effect on quality
- More quantization levels reduce error and improve signal quality, but they also increase the number of bits needed to represent each sample.
Example
- If a signal is quantized into 8 levels, each sample is approximated to one of those 8 levels, not to its exact value.
Quantization illustration
Original sample value: •
Nearest quantization level: ─────
Difference = quantization error
3. Encoding
Meaning of encoding
- Encoding is the process of converting the quantized level into a binary number.
Binary representation
- If there are quantization levels, the number of bits per sample is:
Why encoding is important
- It makes the signal suitable for storage, processing, encryption, and transmission over digital networks.
Bit stream formation
- After encoding each sample, all binary words are arranged sequentially to form the PCM signal.
Example
- If there are 8 quantization levels, then bits are needed per sample because .
Common application
- In digital telephony, speech is sampled, quantized, and encoded into binary code before being transmitted.
Encoding illustration
Quantized level 1 -> 000
Quantized level 2 -> 001
Quantized level 3 -> 010
Quantized level 4 -> 011
...
Binary words combine to form the PCM bit stream
Working / Process
1. The analog input signal is first band-limited and then sampled at regular intervals
- Before conversion, the signal is often passed through an anti-aliasing filter to remove frequency components above the allowable range.
- Sampling produces discrete-time values from the continuous waveform.
- The chosen sampling rate must satisfy the Nyquist criterion to avoid information loss.
2. Each sample is quantized to the nearest discrete amplitude level
- The sampled value is compared with predefined quantization levels.
- It is rounded to the closest level.
- The rounding difference creates quantization noise, which is unavoidable in PCM unless an infinite number of levels is used.
3. The quantized values are encoded into binary form and transmitted/stored
- Each level is assigned a unique binary code.
- The binary codes are serialized into a PCM bit stream.
- At the receiver, the process is reversed: decoding, digital-to-analog conversion, and smoothing reconstruction are performed to recover the signal.
PCM processing chain
Analog Signal
↓
Anti-aliasing Filter
↓
Sampling
↓
Quantization
↓
Encoding
↓
PCM Bit Stream
↓
Transmission / Storage
↓
Decoding
↓
Reconstruction Filter
↓
Recovered Analog Signal
Advantages / Applications
High noise immunity
- Since PCM uses binary digits, it can withstand noise better than analog transmission. Even if small noise is added, the receiver can usually distinguish 0s and 1s correctly.
Easy storage and processing
- PCM signals can be easily stored in computers, solid-state memory, disks, and digital systems, and they can also be compressed, encrypted, or corrected using error-control methods.
Better long-distance communication
- Digital regeneration techniques can restore PCM signals at repeaters without amplifying noise, making PCM suitable for long communication links.
Wide use in speech and audio
- PCM is used in digital telephony, audio CDs, VoIP systems, voice recording, and multimedia communication.
Compatibility with modern networks
- Since almost all modern communication systems are digital, PCM fits naturally into digital transmission frameworks.
Basis for advanced techniques
- Many digital signal processing and communication methods are built on PCM principles.
Summary
- PCM converts an analog signal into digital form using sampling, quantization, and encoding.
- It is widely used because it gives better noise resistance and works well in digital systems.
- Important terms to remember: sampling, Nyquist theorem, quantization, quantization noise, encoding, bit stream, aliasing