Why We Modulate

Every radio transmission you have ever heard — every voice contact on 40 meters, every FM broadcast on your car radio, every digital FT8 signal spotted on a waterfall — shares one fundamental property: information has been impressed onto a high-frequency carrier wave before being transmitted. That process is called modulation, and this lesson explains from first principles exactly why it is necessary, what it achieves, and why different modulation techniques exist for different purposes.

The Problem with Audio Frequencies

When you speak into a microphone, the microphone converts the pressure waves of your voice into a varying electrical voltage. Human speech occupies roughly 300 Hz to 3,000 Hz — that is, the voltage varies between 300 and 3,000 times per second. Music spans a wider range, from around 20 Hz up to 20,000 Hz. These frequencies are what your ears hear and your brain interprets as sound.

A natural question is: why not connect the microphone output directly to an antenna and radiate those audio-frequency signals as electromagnetic waves? The physics of antennas makes this completely impractical. An antenna works most efficiently when its physical length is a significant fraction of the wavelength of the signal it is trying to radiate — the most common design is a half-wavelength or quarter-wavelength antenna. The wavelength of any electromagnetic wave is determined by the speed of light and the frequency:

At audio frequencies, this formula produces absurd antenna lengths. At 1,000 Hz (1 kHz) — a typical mid-range audio tone — the wavelength is:

No structure of that size could ever be built. Even at 20,000 Hz (20 kHz), the upper limit of human hearing, the quarter-wave antenna length is still 3.75 kilometers (2.3 miles). These are simply not practical lengths. Audio signals, as electrical energy, can travel perfectly well through wires — but through free space as electromagnetic radiation, they are utterly impractical to radiate efficiently.

There is a second problem beyond antenna size. At audio frequencies, many different signals would all occupy the same extremely narrow slice of spectrum. Every transmitter on Earth would be competing in the same tiny band below 20 kHz, and there would be no way to tune to one station versus another. The entire concept of channelization — having separate channels for different users — depends on shifting signals to higher frequencies where the spectrum is wider and signals can be separated.

Antenna size comparison: a quarter-wave antenna at 1 kHz audio frequency would need to be 75 km tall. At 14 MHz (20 m HF band) it is just 5.3 m. At 146 MHz (2 m VHF) it shrinks to 51 cm — a practical handheld antenna. Modulation shifts information to frequencies where antennas are buildable.

View Larger

Antenna Size and Wavelength

Understanding why an antenna must be sized to the wavelength of the signal helps you appreciate what modulation solves. An antenna is a resonant structure — it stores energy in its electric and magnetic fields and releases it as electromagnetic radiation most efficiently when its physical dimensions match the electrical wavelength of the signal. A dipole antenna half a wavelength long is resonant at the intended frequency: current flows to a natural maximum at the center and falls to zero at the tips, producing an efficient radiation pattern.

When an antenna is much shorter than a wavelength, it looks almost entirely like a capacitor to the transmission line — it has very high reactance and very low radiation resistance. The signal energy bounces back rather than radiating into space. You can force a short antenna to work by adding inductors to cancel the capacitive reactance (a process called loading), but the radiation resistance remains very low compared to the feed-point impedance, so efficiency suffers badly. At 1 kHz, even a 100-meter antenna is only 0.00033 wavelengths long — an efficiency disaster.

At radio frequencies, wavelengths become physically manageable. In the HF amateur bands (3–30 MHz), wavelengths range from 10 to 100 meters — antennas that fit comfortably in a backyard. On VHF (30–300 MHz), wavelengths are 1 to 10 meters — antennas small enough to mount on a car roof. On UHF (300 MHz – 3 GHz), antennas are centimeters long. The key insight is: by moving information to a higher frequency, we gain antennas that are buildable, efficient, and practical.

The ham radio bands you will work on span a huge range of frequencies and corresponding wavelengths. The table below shows how wavelength and antenna size relate across the most common amateur bands:

Compare any of these practical antenna sizes to what you would need at 1 kHz (75 km for a quarter-wave), and the motivation for modulation becomes immediately clear. By shifting our information signal to ride on a carrier at 7 MHz, for example, we can use a 10-meter dipole — something any licensed ham can erect in a modest yard.

Worked Example: Antenna Lengths Compared

Let's work through the numbers concretely. Suppose you want to transmit a voice signal. The audio content of the voice occupies frequencies from 300 Hz to 3,000 Hz. Compare what happens if you try to transmit at audio frequency versus using modulation to place the signal on a 14.225 MHz carrier (the common 20-meter SSB calling frequency).

This worked example reveals the fundamental benefit of modulation: it translates your information signal from an impractical frequency to a practical one. The information content — your voice, your Morse code, your digital data — is unchanged. Only the vehicle carrying it has changed. That vehicle is the carrier wave.

The Carrier Wave

A carrier wave is a continuous sine wave at a specific radio frequency, produced by an oscillator in the transmitter. By itself, a carrier wave carries no information — it simply oscillates at a fixed frequency with constant amplitude and phase. It is like an empty truck driving down the highway. The truck has the capability to move cargo, but until cargo is loaded, nothing useful is being transported.

The carrier wave has three properties that can be varied to carry information:

• Amplitude — how tall the peaks of the sine wave are
• Frequency — how fast the sine wave completes each cycle
• Phase — where in its cycle the sine wave is at any given instant

If you leave all three properties constant, the wave carries no information. The moment you begin varying one of these properties in a pattern that corresponds to your information signal, you have modulated the carrier. The receiver on the other end can then detect those variations and reconstruct the original information. This is the essence of radio communication.

The carrier frequency is chosen to fall within an amateur band allocation, match the antenna system in use, and suit the propagation conditions for the desired communication path. A 40-meter carrier at 7.074 MHz is well-suited to the FT8 digital mode used for long-distance contacts; a 2-meter carrier at 146.520 MHz is the national simplex calling frequency used for local FM voice contacts. Both carriers serve the same fundamental purpose — providing a radio-frequency vehicle for information — but they differ in frequency to match their intended propagation mode and use case.

The three elements of modulation: a low-frequency audio signal (top) combines with a high-frequency carrier (middle) to produce a modulated signal (bottom) that can be radiated efficiently from a practical antenna.

View Larger

What Modulation Does

Modulation is the process of deliberately varying one or more properties of a carrier wave in accordance with an information signal. The information signal is called the modulating signal or baseband signal. The result is the modulated signal — a radio-frequency wave whose variations encode the original information.

From a spectral standpoint, modulation transforms the information signal in a very specific way. A single audio tone at frequency f_m placed on a carrier at frequency f_c does not simply shift the audio tone up in frequency. Instead, it creates new frequency components called sidebands that appear at f_c + f_m and f_c − f_m. These sidebands, sitting symmetrically above and below the carrier, are what actually carry the information. The carrier itself, in many modulation schemes, contains no information at all — it is purely a reference frequency.

This sideband structure is crucial to understanding why all radio signals occupy a range of frequencies (a bandwidth) rather than a single point on the dial. A voice signal spanning 300 to 3,000 Hz, modulated onto a carrier, produces sidebands spanning 2.7 kHz wide on each side of the carrier. That 2.7 kHz bandwidth is why AM stations on the medium wave broadcast band are spaced 10 kHz apart, and why SSB voice contacts on the HF amateur bands occupy a 2.4–2.8 kHz slice of spectrum. Every detail of how sidebands form, and how receivers extract the information from them, will be covered in the individual lessons on AM, SSB, and FM that follow this one.

The receiver in a radio station performs demodulation — the reverse process of modulation. It takes the received modulated signal, extracts the variations that encode the information, and reconstructs the original audio or data signal. A properly designed receiver tuned to the carrier frequency can recover the information signal even after it has propagated thousands of miles and arrived at the antenna as an incredibly weak signal, often far below one microvolt. The combination of an efficient carrier frequency, a practical antenna, and a sensitive receiver makes global radio communication possible.

Three Parameters You Can Vary: Amplitude, Frequency, Phase

Every form of modulation — from the earliest spark-gap Morse code to the most sophisticated modern digital mode — works by varying the amplitude, the frequency, or the phase of a carrier wave (or some combination of these). Understanding the difference between these three fundamental approaches is the foundation for everything else in this module.

Amplitude Modulation (AM)

In amplitude modulation, the frequency and phase of the carrier remain constant, but the amplitude (peak voltage) of the carrier is made to vary in proportion to the instantaneous value of the modulating signal. When the audio waveform is at its peak positive value, the carrier amplitude is at its maximum. When the audio is at zero, the carrier amplitude returns to its unmodulated value. When the audio waveform dips negative, the carrier amplitude decreases below its unmodulated value. The result is a carrier whose envelope — the outline traced by the peaks of the RF waveform — mirrors the shape of the audio signal exactly.

AM was the first practical modulation technique for voice radio and is still used today in the HF shortwave bands, the aviation communication band (118–137 MHz), and the medium wave AM broadcast band (530–1700 kHz). It is straightforward to demodulate with very simple circuits — even a single diode and capacitor can recover audio from an AM signal — but it is relatively wasteful of power and spectrum compared to more modern techniques.

Frequency Modulation (FM)

In frequency modulation, the amplitude and phase of the carrier remain constant, but the instantaneous frequency of the carrier is made to shift above and below its center frequency in proportion to the amplitude of the modulating signal. Loud audio causes large frequency swings (high deviation); quiet audio causes small frequency swings (low deviation). A positive audio voltage shifts the carrier frequency upward; a negative audio voltage shifts it downward. The amount of frequency shift from the center frequency is called the deviation, and it is the key parameter that determines FM signal bandwidth and quality.

FM is used for VHF and UHF amateur voice communications (the 2-meter and 70-cm bands), the FM broadcast band (88–108 MHz), and many commercial radio services. FM has significantly better noise rejection than AM in strong-signal conditions — a property called the capture effect — making it the preferred mode for repeater systems and local communications where signal levels are generally strong.

Phase Modulation (PM)

In phase modulation, the amplitude and frequency of the carrier remain nominally constant, but the phase of the carrier shifts in proportion to the instantaneous value of the modulating signal. At first glance, frequency modulation and phase modulation sound very similar — and they are closely related. In fact, FM can be produced by integrating the audio signal and then applying it as phase modulation. This relationship is exploited in many modern transmitter designs, particularly in synthesizer-based radios where the VCO naturally acts as a phase modulator when its input is an integrated audio signal.

Digital phase modulation — where the phase is switched between discrete states rather than varied continuously — underlies most modern digital communications. Binary Phase Shift Keying (BPSK), Quadrature Phase Shift Keying (QPSK), and the related 8PSK and 16QAM schemes used in digital modes like PSK31, FT8, and JS8Call all use phase shifts to encode bits. These digital PM techniques achieve very high information density in a narrow bandwidth — FT8, for example, completes a full exchange in 15 seconds using only 50 Hz of bandwidth.

The three fundamental modulation types compared: AM varies the carrier amplitude, FM varies the carrier frequency, and PM varies the carrier phase. All three use the same carrier frequency but encode information differently. The top trace is the original audio modulating signal in all three cases.

View Larger

Overview of Modulation Types

Beyond the three fundamental analog modulation types, a wide variety of modulation schemes have been developed for specific applications in amateur radio and commercial communications. This lesson provides an overview; each type is covered in detail in the lessons that follow.

Looking at this table, you can already see the enormous range of bandwidths involved — from 31 Hz for PSK31 to 180 kHz for FM broadcasting. Bandwidth is not an accident; it is directly determined by the modulation method and the information content being carried. A lesson entirely dedicated to bandwidth comparisons (M11I) follows later in this module. For now, appreciate that the choice of modulation mode is a tradeoff between bandwidth (how much spectrum you use), power efficiency, noise immunity, and simplicity of equipment.

The Electromagnetic Spectrum and Amateur Allocations

The electromagnetic spectrum extends from zero frequency upward through radio waves, microwaves, infrared light, visible light, ultraviolet, X-rays, and beyond. Radio waves occupy frequencies from roughly 3 kHz to 300 GHz, a span of eight orders of magnitude. This vast range is divided into named bands by the International Telecommunication Union (ITU), and within those bands, various services — amateur radio, broadcasting, aviation, marine, military, commercial — are allocated specific frequency ranges.

Amateur radio in the United States is regulated by the FCC (Federal Communications Commission) and is allocated frequencies across nearly the entire radio spectrum, from 135.7 kHz on the long-wave band all the way up to frequencies above 300 GHz in the millimeter wave range. Each band has its own propagation characteristics, practical antenna sizes, typical modulation modes, and operating culture. Understanding which modes are used on which bands — and why — is part of what separates an effective operator from one who merely holds a license.

Notice how modulation modes shift with frequency. On the HF bands, where sky-wave propagation allows contacts with the entire world, voice operators use SSB because it concentrates transmitter power into a single sideband, maximizing the chance of being heard at great distances. On VHF and UHF, where signals travel by line of sight and repeaters provide strong local coverage, FM dominates because its excellent noise rejection makes it highly intelligible even in mobile environments with signal fading. CW and digital modes appear on both HF and VHF, valued for their narrow bandwidth and ability to work in conditions where voice modes fail entirely.

Why Different Modes for Different Purposes

No single modulation mode is best for all situations. Each mode involves a tradeoff between power efficiency, bandwidth, complexity, noise immunity, and the nature of the information being transmitted. Understanding these tradeoffs is what allows you to choose the right mode for the job.

Power Efficiency

A 100 W AM transmitter puts most of its power into the carrier, with only a fraction in each sideband — the carrier itself carries no information. An SSB transmitter puts all 100 W into one sideband where the information actually lives. This gives SSB a 9 dB theoretical advantage over full carrier AM, meaning an SSB signal sounds as loud as an AM signal from a transmitter nine times more powerful. For DX operators trying to reach across the Pacific with 100 W and a wire antenna, that 9 dB matters enormously. On the other hand, FM puts all its power into the carrier at all times, which is power-efficient but the information is in the frequency variations rather than amplitude, so power efficiency comparisons with AM and SSB are not directly applicable in the same way.

Noise Immunity

Noise in radio — from lightning, power lines, car ignition systems, and electronic equipment — tends to affect the amplitude of received signals. FM, which encodes information in frequency rather than amplitude, is largely immune to amplitude noise. A limiter circuit in the FM receiver strips away amplitude variations before demodulation, and with them goes most of the noise. This is why FM sounds so much cleaner than AM for local strong-signal communication, and why the VHF bands use FM almost exclusively for voice. However, once an FM signal drops below the threshold where the receiver can track the frequency, it falls off a cliff — intelligibility suddenly collapses completely. AM degrades more gracefully as signals weaken.

Bandwidth and Spectrum Efficiency

Available spectrum is finite and shared. A mode that uses less bandwidth allows more stations to operate simultaneously in a given piece of spectrum. CW occupies 150–500 Hz. PSK31 occupies 31 Hz — in the bandwidth of a single CW signal, you could fit several dozen PSK31 contacts. FT8 is narrower still. SSB is more efficient than AM because it transmits only one sideband and suppresses the carrier. FM is less spectrum-efficient than SSB for voice, but the FM broadcast band (with its high-quality stereo audio and 180 kHz bandwidth per station) is a deliberate design choice trading bandwidth for audio quality. In the crowded HF bands, spectrum efficiency is paramount, which is why SSB displaced AM for HF voice decades ago.

Equipment Complexity

Simple circuits — a diode, a capacitor, and a resistor — can demodulate an AM signal. CW requires nothing more than turning a carrier on and off. SSB requires more sophisticated circuitry to suppress the carrier and unwanted sideband in the transmitter, and a product detector for demodulation. FM requires a discriminator or ratio detector, or in modern radios, a software-defined demodulator. Digital modes require a computer (or dedicated DSP chip) to generate and decode their complex waveforms. Historically, equipment complexity constrained which modes were practical; today, software-defined radios have made every modulation type equally accessible, because the demodulation happens in software rather than discrete hardware.

The Information Bandwidth Match

There is a fundamental principle called Shannon's theorem that links the maximum information rate a channel can carry to its bandwidth and signal-to-noise ratio. Modes that use very narrow bandwidth must either carry less information per second (like CW and PSK31, which are slow text modes) or use very sophisticated error-correction coding to squeeze more bits through fewer hertz. FT8 achieves remarkable weak-signal performance precisely because its designers accepted a very low information throughput — a full four-field exchange (call signs, signal report, and acknowledgment) takes 75 seconds — in exchange for the ability to make contacts at signal levels 15 dB below what SSB voice requires. Every modulation mode represents a specific set of choices along these tradeoffs.