Amplitude modulation (AM) has an inherent weakness: noise affects amplitude directly, degrading the signal. Frequency modulation takes a different approach — it encodes information in the frequency of the carrier, not its amplitude. The result is a modulation scheme dramatically more resistant to noise, at the cost of wider bandwidth.
In AM, the carrier's amplitude rises and falls in proportion to the message signal. In FM, the carrier's amplitude stays constant, but its instantaneous frequency varies in proportion to the message. Louder audio means a higher frequency deviation; quieter audio means less deviation.
Intuition: Think of FM like a speedometer. The message signal is the speed — as it increases, the needle (carrier frequency) deflects more. The amplitude of the needle's movement doesn't change; only where it points does.
The mathematical description of an FM signal captures both the constant amplitude and the frequency that varies with the integral of the message:
The key insight is in the argument of the cosine. Instead of just 2πf_c·t (a fixed-rate angle sweep), we add 2πk_f·∫x(τ)dτ — a phase term that grows faster when x is large and slower when x is small. This is what causes the frequency to vary.
The instantaneous frequency of the FM signal at any moment is the rate of change of the total phase. If the total phase is θ(t) = 2πf_c·t + 2πk_f·∫x(τ)dτ, then:
The frequency deviation Δf is the maximum swing in instantaneous frequency from the carrier. For a sinusoidal message x(t) = A_m·cos(2πf_m·t):
The modulation index β (beta) for FM is defined as the ratio of frequency deviation to message frequency:
The modulation index β determines whether an FM signal is narrowband or wideband:
β ≪ 1 (typically β < 0.3). Bandwidth ≈ 2f_m — similar to AM. Used in two-way radio and marine VHF. (Civil aviation voice is AM, not FM — see M3-L2.) Lower noise immunity than wideband FM.
β ≫ 1 (typically β = 2–10). Much wider bandwidth but dramatically better noise immunity. Used in broadcast FM radio, high-fidelity audio transmission.
Unlike AM where the spectrum is simply carrier ± sidebands, the FM spectrum is mathematically more complex. For a single-tone message, the FM signal generates an infinite series of sidebands spaced at multiples of f_m from the carrier. The amplitude of each sideband is given by Bessel functions of the first kind, J_n(β):
Bessel function property: As β increases, energy spreads from the carrier into more sideband pairs. At certain values of β (β ≈ 2.4, 5.5, 8.6...), J_0(β) = 0 — the carrier disappears entirely! All the power is in the sidebands. This has no AM analogue.
The fundamental advantage of FM over AM is noise immunity. Atmospheric noise, electrical interference, and other disturbances typically add to the amplitude of a signal. In AM, this directly corrupts the demodulated audio. In FM, the information is in the frequency, not the amplitude — so amplitude noise is largely irrelevant.
FM receivers use limiters — circuits that clip the signal to a constant amplitude before demodulation. Any amplitude variation (noise) is stripped away, leaving only the frequency variation that carries the audio. This is the origin of FM's superior sound quality.
The capture effect: When two FM stations transmit on the same frequency, the stronger one completely dominates — the weaker is suppressed. In AM, both are heard together (co-channel interference). FM's capture effect is both a strength (cleaner reception) and a limitation (no graceful degradation).