Digital Modulation Overview

Digital modulation is the process of mapping discrete binary data onto a continuous carrier signal for transmission over a physical channel. Every modern communication system, from mobile networks to satellite links, depends on a chosen modulation scheme that balances power efficiency, bandwidth usage, and noise robustness. Understanding the core trade-offs in digital modulation is the first step toward analyzing any digital communication system.

Core Concept: What Digital Modulation Does

A carrier signal is a high-frequency sinusoid described by three parameters: amplitude, frequency, and phase. Digital modulation selectively varies one or more of these parameters in synchrony with the incoming bit stream. Each distinct combination of parameter values represents a symbol, and the set of all possible symbols forms the constellation diagram of the scheme. The receiver must identify which symbol was transmitted by observing the received (noisy) signal.

In binary schemes, one bit maps to one symbol. In M-ary schemes, log₂(M) bits map to one symbol. Increasing M improves spectral efficiency (bits per second per Hz) but requires closer symbol spacing, making the system more sensitive to noise. This core trade-off between spectral efficiency and bit error rate (BER) drives the selection of a modulation scheme for any real system.

Coherent vs Non-Coherent Detection

In coherent detection, the receiver generates a local carrier that is phase-synchronized with the transmitted carrier. The received signal is multiplied by this reference and integrated (correlator receiver). Coherent detection achieves the minimum possible BER for a given Eb/N0, but requires carrier phase recovery circuitry, which adds complexity.

In non-coherent detection, the receiver does not need to know the carrier phase. Envelope detection or differential detection is used instead. This simplifies hardware but degrades BER performance. For BFSK, the coherent receiver performs about 3 dB better than its non-coherent counterpart at the same BER. DPSK is a popular non-coherent variant of PSK used in practice.

Power Efficiency vs Bandwidth Efficiency

Power efficiency measures how much signal energy per bit (Eb) is required to achieve an acceptable BER. A power-efficient scheme achieves low BER at low Eb/N0. Bandwidth efficiency (η) is defined as the ratio of bit rate Rb to bandwidth B, expressed in bits/s/Hz. These two metrics are generally in conflict: increasing spectral efficiency requires denser constellations, which demand higher Eb/N0.

The Shannon-Hartley theorem gives the theoretical upper bound: C = B log₂(1 + SNR). Practical schemes are plotted on the bandwidth-power efficiency plane, and the goal is to approach the Shannon limit. BPSK and BFSK favor power efficiency; QPSK offers a balance; higher-order QAM sacrifices power efficiency for spectral efficiency.

Mathematical Expression

The general digitally modulated signal is written as s(t) = A(t) cos(2π f(t) t + φ(t)), where A(t), f(t), and φ(t) are the instantaneous amplitude, frequency, and phase respectively, each controlled by the symbol being transmitted. Bandwidth efficiency is:

η = Rb / B = log₂(M) / (B × Ts)

where M is the constellation size, Ts is the symbol duration, and B is the transmission bandwidth. BER for coherent BPSK is given by Pb = Q(sqrt(2Eb/N0)), which sets the benchmark for comparing all other schemes.

Practical Understanding

In 4G LTE, QPSK is used in poor channel conditions for reliability while 64-QAM is selected when the SNR is high, achieving up to 6 bits per symbol. The base station continuously adapts the modulation order through adaptive modulation and coding (AMC). In satellite communications where power is scarce, BPSK and QPSK dominate because they are the most power-efficient. In cable TV (DOCSIS), 256-QAM and higher are used because bandwidth is limited but SNR is very high.

Given:
Bit rate Rb = 10 Mbps, Modulation = 16-QAM (M = 16), Bandwidth B = 3 MHz

Why this formula applies:
Bandwidth efficiency is defined as η = Rb / B and theoretically equals log₂(M) for ideal Nyquist signaling.

Formula:
η = Rb / B
Theoretical η = log₂(M)

Substitution:
η = 10 Mbps / 3 MHz = 3.33 bits/s/Hz
Theoretical η = log₂(16) = 4 bits/s/Hz

Calculation:
Actual efficiency = 3.33 bits/s/Hz (below theoretical due to guard bands and excess bandwidth)
Gap from ideal = 4 - 3.33 = 0.67 bits/s/Hz

Final Answer with units:
Bandwidth efficiency = 3.33 bits/s/Hz (practical), 4 bits/s/Hz (theoretical for 16-QAM)

Exam Tip: GATE frequently asks which modulation has the best bandwidth efficiency for a given M. Remember: higher M means more bits/symbol and better spectral efficiency, but requires higher SNR. BPSK has η ≈ 1 b/s/Hz; QPSK doubles it to ≈ 2 b/s/Hz with the same BER performance relative to Eb/N0.

Quick Revision

• Digital modulation varies amplitude (ASK), frequency (FSK), or phase (PSK) of a carrier to encode binary data.
• Spectral efficiency η = Rb/B = log₂(M) bits/s/Hz for M-ary modulation under ideal Nyquist conditions.
• Coherent detection is 3 dB better than non-coherent but requires carrier synchronization.
• Power efficiency and bandwidth efficiency are opposing goals; higher M improves bandwidth efficiency but worsens power efficiency.
• BPSK is the most power-efficient binary scheme: Pb = Q(sqrt(2Eb/N0)).
• Shannon capacity C = B log₂(1 + SNR) gives the theoretical upper bound that no practical scheme can exceed.
• Common trap: increasing M always improves spectral efficiency but never improves BER at the same Eb/N0.