Bandwidth in Communication Systems

Bandwidth is one of the most critical parameters in communication systems. It determines the information-carrying capacity of a channel and directly affects data transmission rates, system cost, and spectral efficiency.

Definition of Bandwidth

Bandwidth has several related but distinct meanings depending on context:

Signal Bandwidth

The range of frequencies contained in a signal:

BW_signal = f_max - f_min

Channel Bandwidth

The range of frequencies that a channel can pass with acceptable attenuation:

BW_channel = f_upper - f_lower (defined at -3dB points)

Transmission Bandwidth

The minimum bandwidth required to transmit a signal without unacceptable distortion.

Types of Bandwidth

Bandwidth of Common Signals

Bandwidth and Data Rate Relationships

Nyquist Theorem (Noiseless Channel)

C = 2B × log₂(M) bits/second

Where:

• C = maximum data rate (channel capacity)
• B = channel bandwidth in Hz
• M = number of discrete signal levels

Shannon-Hartley Theorem (Noisy Channel)

C = B × log₂(1 + SNR) bits/second

Where:

• C = channel capacity (theoretical maximum)
• B = bandwidth in Hz
• SNR = signal-to-noise ratio (linear, not dB)

Bandwidth Efficiency (Spectral Efficiency)

η = R/B bits/s/Hz

Where R = actual data rate, B = occupied bandwidth.

Bandwidth Requirements for Modulation Schemes

Amplitude Modulation (AM)

• DSB-FC: BW = 2W (W = message bandwidth)
• DSB-SC: BW = 2W
• SSB: BW = W
• VSB: BW = W + f_vestige

Frequency Modulation (FM)

Carson's Rule: BW ≈ 2(Δf + W) = 2W(β + 1)

Where:

• Δf = peak frequency deviation
• W = message bandwidth
• β = Δf/W (modulation index)

Digital Modulation

• Minimum BW = R_s = R_b/log₂M (symbol rate)
• Practical BW = R_s(1 + α) where α = excess bandwidth (roll-off factor, 0 to 1)

Bandwidth-Power Trade-off

Key Principle: You can trade bandwidth for power and vice versa:

• Wideband FM uses more bandwidth but achieves better SNR than AM
• Spread spectrum uses wide bandwidth to achieve interference rejection
• Shannon's theorem sets the ultimate limit for any trade-off

Solved Example

Problem: A digital communication link uses 64-QAM modulation with roll-off factor α = 0.25 and must support 100 Mbps. Calculate: (a) symbol rate, (b) minimum bandwidth, (c) actual bandwidth with excess, (d) spectral efficiency.

Solution:

(a) Symbol rate: 64-QAM → M = 64, bits per symbol = log₂(64) = 6 R_s = R_b / log₂(M) = 100 × 10⁶ / 6 = 16.67 MSymbols/s

(b) Minimum bandwidth (Nyquist): BW_min = R_s / 2 = 16.67 / 2 = 8.33 MHz (double-sided: R_s = 16.67 MHz)

Actually for passband: BW_min = R_s = 16.67 MHz

(c) Actual bandwidth with roll-off: BW_actual = R_s(1 + α) = 16.67 × (1 + 0.25) = 20.83 MHz

(d) Spectral efficiency: η = R_b / BW_actual = 100 / 20.83 = 4.8 bits/s/Hz

Interview Questions

Q1: Why is bandwidth considered the most valuable resource in communication?

Bandwidth is finite and shared among all users. Radio spectrum is regulated by governments and licenses cost billions. Unlike power (which can be increased by adding amplifiers), bandwidth cannot be created. Efficient use of bandwidth through better modulation, coding, and multiple access schemes is a primary goal of communication engineering.

Q2: Explain the relationship between Carson's rule and FM bandwidth.

Carson's rule (BW ≈ 2(Δf + W)) gives the approximate bandwidth containing 98% of FM signal power. For narrowband FM (β<<1), BW ≈ 2W (same as AM). For wideband FM (β>>1), BW ≈ 2Δf. The bandwidth increases with modulation index, which is the trade-off for improved noise performance.

Q3: How does Shannon's theorem relate to practical system design?

Shannon's theorem gives the theoretical maximum data rate C = B·log₂(1+SNR). No practical system can exceed this limit. Modern systems with turbo codes or LDPC approach within 0.5 dB of Shannon's limit. The theorem guides engineers in choosing bandwidth and power allocation to maximize throughput.

Q4: What is spectral efficiency and why is it important for 5G?

Spectral efficiency (bits/s/Hz) measures how effectively bandwidth is utilized. 5G targets 30 bits/s/Hz peak spectral efficiency using massive MIMO, 256-QAM, and carrier aggregation. Higher spectral efficiency means more users can be served in limited spectrum, which is critical as mobile data demand grows exponentially while spectrum remains fixed.