Modulation Principles

Every radio broadcast, Wi-Fi packet, and phone call you make travels through the air as an electromagnetic wave. But a voice or a song is a low-frequency signal — far too slow to radiate efficiently as an antenna of any practical size. The solution is modulation: the process of imprinting a low-frequency information signal (the 'baseband' or 'message' signal) onto a high-frequency electromagnetic wave called the carrier. The carrier does the travelling; the message signal shapes how it travels. There are two classic ways to encode information onto a carrier wave. In Amplitude Modulation (AM), the strength (amplitude) of the carrier is varied in proportion to the message signal while the frequency stays fixed. In Frequency Modulation (FM), the frequency of the carrier is varied in proportion to the message while the amplitude stays fixed. Both techniques are still in everyday use — AM covers very large distances cheaply, and FM delivers cleaner audio for local broadcasts because it is inherently less affected by electrical noise. Understanding modulation also requires understanding bandwidth — the range of frequencies a modulated signal occupies. Because modulation creates new frequency components called sidebands around the carrier, every radio station must be allocated its own slice of the frequency spectrum. Bandwidth is therefore not just a technical property; it is a limited public resource managed by government regulators worldwide.

Key takeaways
  • The Carrier and Message Signals
  • Amplitude Modulation (AM)
  • Modulation Index for AM
  • Frequency Modulation (FM)
  • Why FM is Less Noisy

The Carrier and Message Signals

Start with two pure sinusoids. The carrier is c(t) = A_c cos(2πf_c t), where f_c is typically hundreds of kHz (AM) or tens of MHz (FM). The message is m(t) = A_m cos(2πf_m t), where f_m might be a few kHz — a note in a song. Alone, m(t) has too low a frequency to broadcast; alone, c(t) carries no information. Modulation combines them cleverly.

Amplitude Modulation (AM)

In AM the carrier's amplitude is made to track the message. The transmitted signal is s_AM(t) = A_c [1 + k_a m(t)] cos(2πf_c t), where k_a is the amplitude sensitivity. If m(t) = A_m cos(2πf_m t), using the product-to-sum identity cos(A)cos(B) = ½[cos(A−B) + cos(A+B)], the AM signal contains three frequency components: the original carrier at f_c, a lower sideband at f_c − f_m, and an upper sideband at f_c + f_m. The total bandwidth occupied is B_AM = 2f_m — twice the highest message frequency. For an audio channel up to 5 kHz, AM uses 10 kHz of spectrum.

Modulation Index for AM

The AM modulation index is μ = k_a A_m = A_m / A_c (when k_a = 1/A_c). It must satisfy 0 ≤ μ ≤ 1; if μ > 1 the carrier becomes 'over-modulated', the envelope distorts, and the signal cannot be correctly demodulated by a simple envelope detector. At μ = 1 (100% modulation), the signal amplitude swings between 0 and 2A_c.

Frequency Modulation (FM)

In FM the instantaneous frequency deviates from f_c in proportion to the message: f_i(t) = f_c + k_f m(t), where k_f is the frequency sensitivity in Hz/V. The transmitted signal is s_FM(t) = A_c cos[2πf_c t + 2πk_f ∫m(τ)dτ]. For a sinusoidal message, the FM modulation index is β = k_f A_m / f_m = Δf / f_m, where Δf is the peak frequency deviation. FM generates an infinite series of sidebands spaced f_m apart, but most energy falls within Carson's Rule bandwidth: B_FM ≈ 2(Δf + f_m) = 2f_m(β + 1). Standard broadcast FM uses Δf = 75 kHz and audio up to 15 kHz, giving B_FM ≈ 2(75 + 15) = 180 kHz — about 18 times wider than AM.

Why FM is Less Noisy

Electrical noise (lightning, motors, appliances) primarily adds random amplitude variations to a signal — exactly what AM tries to encode as information. An FM receiver uses a limiter circuit that clips any amplitude fluctuations before demodulation, throwing away the noise along the way. FM therefore provides a much better signal-to-noise ratio, at the cost of using more bandwidth. This trade-off — bandwidth for noise immunity — is a fundamental theme in all of communications engineering.

The Key Equation

BFM≈2(Δf+fm)=2fm(β+1)B_{FM} \approx 2(\Delta f + f_m) = 2f_m(\beta + 1)
  • B_{FM} — FM bandwidth (Hz): The range of frequencies occupied by the FM signal in the spectrum.
  • \Delta f — Peak frequency deviation (Hz): The maximum amount by which the carrier frequency shifts away from its centre value; set by transmitter design.
  • f_m — Maximum message frequency (Hz): The highest frequency component in the audio or data signal being transmitted.
  • \beta — FM modulation index (dimensionless): Ratio Δf / f_m; describes how much the frequency swings relative to the message frequency.

Worked Example

Problem. A broadcast FM station uses a peak frequency deviation of Δf = 75 kHz and transmits audio up to f_m = 15 kHz. (a) Calculate the FM modulation index β. (b) Use Carson's Rule to find the bandwidth of the transmitted signal. (c) How many such FM stations can fit in a 20 MHz block of spectrum if each station must have a guard band of 10 kHz on each side?

β=5,BFM=180 kHz,N=20 MHz200 kHz=100 stations\beta = 5,\quad B_{FM} = 180\text{ kHz},\quad N = \frac{20\text{ MHz}}{200\text{ kHz}} = 100\text{ stations}

Answer. β = 5; B_FM = 180 kHz; 100 stations fit in a 20 MHz block.

Figure
Diagram
AM vs FM: Signal Comparison in Time and FrequencyMessage signal m(t)low frequencyAM signal s_AM(t)amplitude varieswith messageFM signal s_FM(t)frequency varieswith messageamplitude constantFrequency Spectrum (Sidebands)f_cf_c-f_mf_c+f_mAM: 3 lines, B=2f_mf_cFM: many sidebands, wider B
Origins · 1 minEdwin Howard Armstrong and how this was discovered
EH
Edwin Howard Armstrong
1890–1954 · American

Armstrong invented the regenerative circuit (1914), the superheterodyne receiver (1918), and most crucially, wideband Frequency Modulation (FM) radio (1933). His FM system demonstrated dramatically better audio quality and noise immunity than AM, fundamentally changing broadcasting. He built the first FM radio station atop the Palisades in New Jersey to prove the technology worked in practice. Armstrong spent the last decade of his life in exhausting legal battles defending his FM patents against RCA and others. He tragically died in 1954 before winning most of those suits — his widow ultimately collected settlements from over 20 companies, vindicating his inventions posthumously.

  • Electromagnetic Waves and the Spectrum — Carrier waves in AM and FM are electromagnetic waves; understanding their frequency, wavelength and propagation is essential before studying modulation.
  • Resonance in LC Circuits — Tuning a radio to a station means selecting one carrier frequency using an LC resonant circuit; the same physics underlies both transmitter oscillators and receiver tuners.
  • Superposition and Beats — AM sidebands arise from multiplying two sinusoids — the same product-to-sum trigonometric identity that explains beat frequencies when two notes of similar pitch are played together.
  • Digital Modulation and Sampling — Modern communications use digital variants (ASK, FSK, PSK, QAM) that extend the same AM/FM principles to discrete symbols; the Nyquist sampling theorem sets the bridge from analogue to digital.