EEL4514C Communication Systems and Components, Fall 2026, Lecture 15

Overview

Mingyue Ji explains why conventional AM adds a carrier to DSB-SC: the carrier enables simpler demodulation without exact phase synchronization, at the cost of power efficiency. Under zero-DC message and carrier-frequency assumptions, AM power efficiency is Pm/(A² + Pm), reaching only 33% for a fully modulated single tone; square-law demodulation is simple but requires fc > 2B to separate its doubled-bandwidth components.

Key takeaways

Coherent DSB-SC detection depends on carrier-phase alignment: multiplying by a mismatched cosine can produce a zero or inverted message, so carrier recovery or a PLL is needed.
Conventional AM uses μ = mp/A ≤ 1 to preserve a nonnegative envelope; μ = 1 is 100% modulation, while μ > 1 causes envelope inversion and distortion.
With zero message DC and fc > B, the carrier and DSB-SC component are orthogonal, giving total AM power (A² + Pm)/2 and efficiency Pm/(A² + Pm).
For single-tone AM, efficiency is μ²/(2 + μ²); even at maximum undistorted modulation, μ = 1, only about 33% of transmitted power carries message information.
Square-law demodulation squares, low-pass filters, square-roots, DC-blocks, and rescales the signal, but its frequency expansion forces the stricter carrier condition fc > 2B.

Chapters

0:00 DSB-SC Coherent Detection and the Carrier Phase Problem

Multiplying a received DSB-SC signal by a cosine at the carrier frequency creates a baseband copy of m(t) plus high-frequency terms.
A low-pass filter removes the high-frequency terms, and a gain correction restores the message amplitude.
A phase mismatch can make coherent detection recover no signal or the inverted message, motivating carrier recovery or a phase-locked loop (PLL).

3:17 AM Adds a Carrier to Enable Envelope-Based Recovery

Conventional AM adds a carrier term to the DSB-SC signal, producing the form (A + m(t))cos(ωct).
The envelope follows A + m(t), allowing recovery without reproducing the received carrier's exact phase.
For an undistorted envelope, A + m(t) must remain positive; with mp denoting the message's peak magnitude, the modulation index is μ = mp/A ≤ 1.

5:42 AM Modulation Index: 100% Modulation and Overmodulation

At μ = 1, the AM signal is 100% modulated: its envelope just reaches zero at its minimum.
When μ > 1, the signal is overmodulated and the envelope crosses zero, so simple envelope recovery becomes distorted.
Mingyue Ji introduces power efficiency and AM demodulation as the lecture's two main topics.

9:04 AM Power Efficiency and the Message-Signal Assumptions

AM power efficiency is the useful message-bearing power divided by total transmitted power, expressed as a percentage.
The derivation assumes message bandwidth B, carrier frequency fc > B, and zero DC content in m(t).
A coupling capacitor can remove a constant DC offset from the message before modulation.
Expanding the AM signal's power produces carrier power, a carrier–message cross term, and modulated-message power.

16:05 Parseval’s Theorem Shows the Carrier and DSB-SC Terms Are Orthogonal

The cross term between the carrier and m(t)cos(ωct) is evaluated in the frequency domain using the Fourier-transform inner-product relationship.
The cosine spectrum consists of impulses at ±fc; multiplying by the shifted message spectra samples m(f) at 0 and ±2fc.
Zero message DC and fc > B make those sampled terms vanish, so the carrier and DSB-SC component are orthogonal.
The modulated-message power is Pm/2, since frequency shifting preserves power and the two sidebands each contribute a factor of one-half.

31:26 AM Total Power and the Efficiency Formula

With carrier power A²/2 and message-bearing power Pm/2, total AM power is (A² + Pm)/2.
The resulting efficiency is ηAM = [Pm/(A² + Pm)] × 100%, under the stated zero-DC and fc > B assumptions.
The carrier consumes power but carries no message information, so increasing the carrier amplitude reduces the fraction of power devoted to the message.

35:32 Single-Tone AM Reaches Only 33% Efficiency at Full Modulation

For a single-tone message with peak amplitude μA, message power is Pm = μ²A²/2.
Substitution gives tone-modulation efficiency ηAM = [μ²/(2 + μ²)] × 100%.
At μ = 1, efficiency is 1/3, or about 33%; at μ = 1/2 it is about 11.1%, and at μ = 1/4 it is about 3.0%.
AM remains useful because its demodulation can be much simpler, not because it uses transmitted power efficiently.

39:31 Square-Law AM Demodulation Uses Filtering, a Square Root, and DC Blocking

The proposed detector squares the received AM signal, yielding (A + m(t))²cos²(ωct).
Using cos²(ωct) = [1 + cos(2ωct)]/2 and low-pass filtering removes the component around 2fc, leaving (A + m(t))²/2.
When μ ≤ 1, taking the square root produces (A + m(t))/√2; a DC blocker removes the A term, and scaling recovers m(t).

45:57 Why Square-Law Detection Requires a Higher Carrier Frequency

Squaring in time corresponds to convolution in frequency, so the message-related baseband bandwidth expands from B to 2B.
The squared signal also has components around 2fc, extending from 2fc − 2B to 2fc + 2B.
To keep that band separate from the baseband through filtering, the condition is 2fc − 2B > 2B, or fc > 2B.
Compared with fc > B in the earlier power analysis, square-law detection requires a higher carrier frequency and a wider low-pass filter.

Summary, takeaways, and chapters were generated by AI from the video's transcript and may contain errors. The video belongs to its creator, Mingyue Ji.