UNIT 5

Digital Modulation Techniques

Simple Notes with Diagrams, Flowcharts & Equations
1. Introduction
Digital modulation is the process of changing a property of a high-frequency carrier signal (amplitude, frequency or
phase) according to a digital (binary) message signal, so that it can be transmitted over long distances using antennas
or cables.
In simple words: we take 0s and 1s of data and 'hide' them inside a carrier wave by changing its amplitude,
frequency, or phase.

Why do we need digital modulation?

• Digital signals cannot travel long distances directly (they get weak/noisy).
• A carrier wave (high frequency) can travel far and needs a smaller antenna.
• Digital modulation gives better noise immunity than analog modulation.
• It allows multiplexing (sending many signals at once) and secure transmission.

Types of Digital Modulation

There are 3 basic types, based on which property of the carrier is changed:
• ASK — Amplitude Shift Keying (amplitude changes)
• FSK — Frequency Shift Keying (frequency changes)
• PSK — Phase Shift Keying (phase changes)

Fig 1.1: Flowchart — Classification of Digital Modulation Techniques

2. Amplitude Shift Keying (ASK)
In ASK, the amplitude of the carrier is changed according to the binary data. Binary '1' is sent as carrier present (full
amplitude) and binary '0' is sent as carrier absent (zero amplitude). This is also called On-Off Keying (OOK).

Equation
S(t) = A·cos(2π fc t) for bit 1
S(t) = 0 for bit 0

Waveform

Fig 2.1: Message signal, Carrier signal, and ASK modulated waveform

Features
• Simplest and cheapest technique to implement.
• Very sensitive to noise, because noise also changes amplitude.
• Bandwidth required = 2 × (bit rate).
• Used in low-cost systems like fibre optics / remote controls.

3. Binary Phase Shift Keying (BPSK)

In BPSK, the phase of the carrier is shifted by 180° to represent binary 1 and 0. Amplitude and frequency stay the
same — only the phase changes.

Equation
S1(t) = A·cos(2π fc t) for bit 1 (phase = 0°)
S2(t) = A·cos(2π fc t + π) for bit 0 (phase = 180°)
Waveform

Fig 3.1: BPSK waveform (top) and BFSK waveform (bottom) compared with data

BPSK Transmitter (Generation)

A balanced modulator multiplies the incoming binary data (as +1/-1 levels) with the carrier signal. When data = 1,
carrier passes as it is; when data = 0, carrier is inverted (180° shift).

Fig 3.2: Block diagram of BPSK Generation

BPSK Receiver (Coherent Detection)

At the receiver, the incoming signal is multiplied by a locally generated carrier (must be exactly synchronized in
phase — hence 'coherent'), passed through a Low Pass Filter to remove high frequency terms, and then a decision
device decides whether the bit is 1 or 0.
 Fig 3.3: Block diagram of BPSK Coherent Detection

Features
• More noise-resistant than ASK and FSK.
• Bandwidth required = 2 × (bit rate).
• Needs a synchronized carrier at receiver (complex hardware).

4. Differential Phase Shift Keying (DPSK)

DPSK is a modified version of BPSK that removes the need for a synchronized carrier at the receiver. Instead of
comparing the phase to a fixed reference, DPSK compares the phase of the current bit with the phase of the previous
bit.
• If the phase does NOT change from the previous bit → data bit is '1' (or '0', depending on convention).
• If the phase DOES change from the previous bit → data bit is the opposite.

Transmitter Block Diagram

The incoming data is first passed through a differential encoder (uses current bit and the previous encoded bit
through an XNOR/XOR logic with 1-bit delay feedback), and then modulated normally like BPSK.

Fig 4.1: Block diagram of DPSK Transmitter

Features
• No need for a coherent (synchronized) carrier at the receiver — simpler receiver.
• Slightly more prone to errors than ideal coherent BPSK (errors can propagate to next bit).
• Used where carrier recovery is difficult, e.g. mobile/fading channels.
5. Quadrature Phase Shift Keying (QPSK)
QPSK sends 2 bits at a time (called a 'dibit') using 4 different phases of the carrier, each 90° apart: 45°, 135°, 225°,
315°. Because 2 bits are sent per symbol, QPSK is twice as bandwidth-efficient as BPSK for the same bit rate.

Equation
S(t) = A·cos(2π fc t + θn), where θn = 45°, 135°, 225°, 315° (n = 1,2,3,4)

Constellation Diagram

Fig 5.1: QPSK Constellation — 4 points, 90° apart, each representing 2 bits

QPSK Transmitter Block Diagram

Serial data is split into two parallel streams (I-channel and Q-channel) by a Serial-to-Parallel converter. The I-
channel is multiplied by cos(carrier) and the Q-channel is multiplied by sin(carrier) (90° phase-shifted carrier). Both
are added together (summer) to give the final QPSK signal.

Fig 5.2: Block diagram of QPSK Generation

Offset QPSK (OQPSK) vs Non-Offset (Normal) QPSK

Non-Offset (Normal) QPSK Offset QPSK (OQPSK)
I and Q bit streams change at the same time instant. Q-channel data is delayed by half a bit-period (Tb/2)
 Non-Offset (Normal) QPSK Offset QPSK (OQPSK)
with respect to I-channel.
Phase can jump suddenly by 180° (both I & Q Phase changes are limited to a maximum of 90° at a
change together). time — smoother transitions.
Causes larger envelope fluctuations after filtering. Causes smaller envelope fluctuations — better for
non-linear amplifiers.
Simple to implement. Slightly more complex (needs an offset/delay circuit)
but more practical for real systems.

6. M-ary PSK
M-ary PSK is the general form of PSK where M = 2^n different phases are used to represent n bits per symbol (M is
the number of possible symbols). BPSK is 2-ary PSK (M=2, n=1) and QPSK is 4-ary PSK (M=4, n=2). Common
higher forms are 8-PSK (M=8, n=3) and 16-PSK (M=16, n=4).

Equation
S(t) = A·cos(2π fc t + 2πk/M), k = 0,1,2,...,M-1

Bandwidth
Bandwidth = 2 × (bit rate) / log2(M)
As M increases, more bits are packed per symbol so bandwidth efficiency improves — but the phase points come
closer together, making it more sensitive to noise.

7. Binary Frequency Shift Keying (BFSK)

In BFSK, the frequency of the carrier is switched between two values: f1 for bit '1' and f2 for bit '0'. Amplitude and
phase stay constant.

Equation
S1(t) = A·cos(2π f1 t) for bit 1
S2(t) = A·cos(2π f2 t) for bit 0

Waveform
(See Fig 3.1 above — bottom waveform shows BFSK with frequency switching between two values.)

Block Diagram
A switch/selector picks one of two oscillators (running at f1 and f2) depending on the incoming bit value, and the
selected signal is sent as output.
 Fig 7.1: Block diagram of BFSK Generation

Features
• More noise-resistant than ASK, less complex than PSK.
• Bandwidth required is larger than ASK/BPSK because two frequencies must be accommodated.
• Can be detected coherently or non-coherently (envelope detection).

8. M-ary FSK
M-ary FSK uses M = 2^n different frequencies to represent n bits per symbol, similar in spirit to M-ary PSK but
changing frequency instead of phase.
Bandwidth = M × (bit rate) / log2(M) (approximately)
M-ary FSK needs more bandwidth than M-ary PSK for the same data rate, but it is simpler to detect (especially non-
coherently) and more robust in some noisy channels.

9. Quadrature Amplitude Modulation (QAM)

QAM combines both ASK and PSK — it changes both the amplitude AND the phase of the carrier at the same time.
This lets it pack even more bits into each symbol than pure PSK, for the same bandwidth.

Equation
S(t) = Ai·cos(2π fc t) + Bi·sin(2π fc t)
where Ai and Bi are the amplitude levels of the in-phase (I) and quadrature (Q) components respectively — different
combinations of Ai and Bi give different constellation points.
Constellation Diagram (16-QAM example)

Fig 9.1: 16-QAM Constellation — 16 points, varying in both amplitude and phase

Features
• Very bandwidth-efficient — widely used in Wi-Fi, cable modems, digital TV.
• More sensitive to noise because both amplitude and phase carry information.
• Higher-order QAM (64-QAM, 256-QAM) sends more bits/symbol but needs a cleaner (higher SNR) channel.

10. Detection of Digitally Modulated Signals

At the receiver side, the modulated signal must be demodulated (detected) to recover the original binary data. There
are two broad approaches:

Fig 10.1: Flowchart — Coherent vs Non-Coherent Detection

 • Coherent Detection: receiver generates a local carrier that is exactly synchronized (same frequency & phase)
with the transmitter's carrier. Gives better performance but hardware is complex. Used for BPSK, QPSK.
• Non-Coherent Detection: no phase synchronization needed; simpler receiver (e.g. envelope detector). Used for
DPSK, non-coherent ASK/FSK — slightly less accurate but much simpler.

11. Summary / Comparison Table

Technique Property Changed Bandwidth (approx.) Noise Immunity
ASK Amplitude 2 × bit rate Poor
BPSK Phase (2 levels) 2 × bit rate Good
DPSK Phase (relative) 2 × bit rate Good (simple RX)
QPSK Phase (4 levels) 1 × bit rate Good
M-ary PSK Phase (M levels) 2×BR / log2(M) Decreases as M
increases
BFSK Frequency (2 levels) > ASK/BPSK Better than ASK
M-ary FSK Frequency (M levels) M×BR / log2(M) Good, needs more BW
approx.
QAM Amplitude + Phase Very efficient Sensitive to noise

12. Quick Revision Points

• ASK → amplitude ON/OFF, simplest, poor noise immunity.
• BPSK → phase shifts by 180°, needs coherent carrier at receiver.
• DPSK → like BPSK, but compares with previous bit's phase — no coherent carrier needed.
• QPSK → 2 bits/symbol, 4 phase points 90° apart; OQPSK offsets Q by Tb/2 for smoother phase transitions.
• M-ary PSK → generalizes PSK to M phases (M = 2^n).
• BFSK → frequency switches between two values for 1 and 0.
• M-ary FSK → generalizes FSK to M frequencies.
• QAM → combines amplitude + phase changes together, most bandwidth-efficient.