DIGITAL COMMUNICATION
23ECPC304
Module-1
Introduction to Digital Communication
Overview of Communication Systems: Comparison between Analog and Digital
Communication; Advantages of Digital Communication; Digital Communication
System Representation Signal Representation and Bandwidth: Time Domain
and Frequency Domain Representation Bandwidth Requirements for Digital
Signals Sampling and Quantization: Nyquist Sampling Theorem; Quantization
Noise and Signal-to Quantization Noise Ratio
Textbook-1: Chapter 1- 1.2
Textbook-2: Chapter 1- 1.3,1.5
TextBook-3: Chapter 7- 7.3,7.8
Analog Communication Digital Communication
Analog communication uses analog signals for the Digital communication uses digital signals for the
transmission of information. transmission of information.
Analog communications are affected by noise. Digital communications are not affected by noise.
The probability of error in analog signals is high. Digital signals are virtually immune to errors.
Digital communication requires less complicated
Analog communication requires complicated hardware.
hardware.
Analog communication is relatively cheaper as compared
Digital communication is costly in nature.
to digital communication.
Low bandwidth is required for analog communication. High bandwidth is required for digital communication.
Digital communication requires comparatively less
High power is required for analog communication.
power.
Analog communication systems are less portable as they Digital communication systems are more portable as
contain heavy and complicated hardware. compared to analog communication systems.
Examples: - Sound shows analog signals Examples: - Computers use digital signals.
Benefits of digital communication over analog communication:
1) Better quality: Digital communication typically provides better quality
and clarity than analog communication, due to its ability to transmit data
without loss or distortion.
2) Greater capacity: Digital signals can carry more information than analog
signals, allowing for greater data transmission and storage.
3) More efficient: Digital communication is faster and more efficient than
analog communication, allowing for quicker transmission of data.
4) Security: Digital signals can be encrypted and protected more easily
than analog signals, providing greater security and privacy for sensitive
information.
MODEL OF A COMMUNICATION SYSTEM
Information sources are classified as analog or discrete. Analog sources, like microphones
or cameras, produce continuous signals, while discrete sources, such as teletypes or
computers, generate sequences of distinct symbols. Analog signals can be converted into
discrete sources via sampling and quantization. Discrete sources are described by
parameters: source alphabet (symbols), symbol rate, alphabet probabilities, and
probabilistic dependence among symbols. These allow construction of probabilistic
models to define source entropy (H) and information rate (R), measured in bits per symbol
or per second. For example, a teletype with 26 letters and six special characters forms a
discrete source alphabet.
A discrete source is described by its alphabet size, symbol rate, probabilities, and
dependencies among symbols. For example, a teletype with 32 symbols operating at 10
symbols per second demonstrates that certain letters occur more frequently, while some
depend on preceding symbols (e.g., Q followed by U). Entropy (H) measures average
information per symbol, expressed in bits per symbol. If all symbols are equally likely,
entropy is maximum; dependencies reduce it. The source information rate (R) equals entropy
multiplied by symbol rate, measured in bits per second. It represents the minimum average
data rate required to transmit source information efficiently.
1. Modulator
● Converts input bit stream into electrical waveforms for transmission.
● Functions:
○ Minimize channel noise.
○ Match transmitted signal spectrum with channel characteristics.
○ Allow multiplexing of multiple signals.
○ Overcome equipment limitations.
● Key parameters: waveform type, duration, power level, bandwidth.
● Noise reduction methods:
○ High signal power and bandwidth.
○ Longer waveform duration (law of large numbers helps average noise).
● Example:
○ At 1 bit/sec, 0 and 1 mapped to cosine signals.
○ Longer duration allows 16 waveforms for 4-bit combinations, but complexity increases.
2. Demodulator
● Reverses modulation, extracting original message from waveforms.
● Key parameter: method of demodulation.
● Influencing factors:
○ Waveform type and duration.
○ Power level at transmitter.
○ Channel noise and distortion.
● Objective: minimize Bit Error Rate (BER).
● BER reduction methods:
○ Improved design of modulator/demodulator.
○ Use of error control coding.
● Together, modulator and demodulator form a modem.
3. Channel Encoder/Decoder
● Purpose: enhance reliability and efficiency of transmission.
● Adds redundancy for error detection/correction.
● Methods:
○ Block Coding: adds rrr error-control bits to kkk information bits.
○ Convolutional Coding: continuously interleaves information and control bits.
● Important parameters:
○ Method of coding.
○ Coding efficiency (input/output data rate ratio).
○ Error control capability.
○ Complexity of implementation.
● Decoder functions:
○ Recover information bits.
4. Other Functional Blocks
● Equalizers: compensate for channel changes.
● Clock Recovery Networks: maintain synchronization between transmitter and receiver.
● Scramblers/Unscramblers: prevent long repetitive sequences (e.g., 0000, 1111) and restore original data.
● Filters: shaping filters at transmitter, receiving filters at receiver.
● A/D Converters: used in digital reception.
5. Data Terminal Equipment (DTE)
● User-end equipment: computer, terminal, etc.
● Functions:
○ Send and receive data from other DTEs.
○ Generate control signals (ready, busy, requests).
○ Ensure flow control of data.
● Connected to modem via standard interfaces (e.g., RS-232C).
● Modem handles waveform conversion; DTE oversees:
○ Error detection.
○ Error correction.
○ Coding tasks.
1Prior to sampling, a low-pass pre-alias filter is used to attenuate those high-frequency
components of the signal that are not essential to the information being conveyed by the
signal.
2. The filtered signal is sampled at a rate slightly higher than the Nyquist rate.
Sampling Process
Sampling is the process of converting a continuous-time signal into
a discrete set of equally spaced values, represented as impulses
weighted by the signal’s amplitude at those instants.
What is a signal?
Think of a signal like a song playing on the radio.
It is smooth and continuous — like a wave that goes on with time.
That is your analog signal
What is sampling?
Imagine you are taking photos of the song wave at equal time intervals.
Instead of drawing the whole smooth wave, you only keep the “dots” you captured.
That process is called sampling.
● If you take a photo every Ts seconds, that time is called the sampling period.
● If you count how many photos you take per second, that is the sampling rate
● (fs=1/Ts ).
How do we show it mathematically?
We use something called the delta function (looks like a vertical arrow).
● Each arrow is placed at the time you took a sample.
● The height of the arrow shows the value of the signal at that instant.
Ideal Sampling Signal:
● Represented as: (7.1)
● This means: multiply each sample value g(nTs) by a Dirac delta
function located at t=nTs.
● The result is a train of impulses, each scaled by the sample value
Simplified Understanding:
● Sampling means taking snapshots of a continuous signal at regular intervals.
● These snapshots are represented mathematically by impulses (delta functions).
● If done properly, the sampled sequence contains all the information needed to reconstruct
the original signal.
Imagine you are filming a running dog.
If you record a video, you see smooth motion (analog).
If you take photos every second, you get snapshots (samples).
More photos (higher sampling rate) → smoother replay.
to a signal when we sample it and look at it in the frequency
domain using the Fourier transform.
Fourier Transform Reminder
● A Fourier transform changes a signal from the time domain (waves/signals
over time) into the frequency domain (which frequencies are present).
What happens when we sample a signal?
(7.2)
● gδ(t): the sampled version of the signal.
● G(f) Fourier transform of the original smooth signal.
● fs: sampling rate (how many samples per second).
when you sample a signal in time, its spectrum (in frequency) gets repeated over and over, spaced by the sampling frequency fs.
So instead of one copy of the spectrum,
you now see infinite copies at multiples of fs. (7.3)
THE QUANTIZATION PROCESS