Also known as: subcarrier, SCA, subsidiary carrier
A subcarrier is a secondary carrier that is modulated with its own information and then placed within the baseband of a larger signal, which is in turn modulated onto the main RF carrier.1 It is a way to stack several independent channels into one transmission by frequency-division multiplexing at baseband — one primary channel near DC and one or more subcarriers at higher baseband frequencies, each demodulated separately at the receiver.
How it works
The transmitter builds a composite baseband: the primary channel occupies the lowest frequencies, and each subcarrier is a tone placed higher up the baseband and modulated (in AM, FM, or a suppressed-carrier scheme) with additional data. That whole composite is then applied to the main RF carrier as if it were ordinary audio. At the receiver, demodulating the main carrier regenerates the composite baseband; a bank of filters then splits out each subcarrier, and each is demodulated in turn. Because a suppressed-carrier subcarrier saves power and spectrum, systems often send only a low-level pilot tone so the receiver can regenerate the missing subcarrier frequency and phase coherently.
The canonical example is the FM broadcast multiplex. Mono audio (L+R) sits from 30 Hz to 15 kHz; a 19 kHz pilot marks a stereo station; the stereo difference signal (L−R) is sent as a double-sideband suppressed-carrier subcarrier centered at 38 kHz (twice the pilot), which the receiver reconstructs by doubling the pilot; and RDS data rides a further subcarrier at 57 kHz (three times the pilot). Additional SCA subcarriers around 67 kHz and 92 kHz have historically carried background music and reading services.
Notice the arithmetic: the 38 kHz stereo subcarrier is exactly twice the 19 kHz pilot and the 57 kHz RDS subcarrier exactly three times it. This is deliberate — deriving every subcarrier from harmonics of one pilot lets the receiver regenerate all of them from a single locked reference, and it keeps the subcarriers from drifting relative to one another. It is the analog ancestor of the coherent-reference thinking that pervades digital systems.
Variants
Subcarriers appear far beyond FM stereo. Analog color television carried the chrominance (color) information on a suppressed subcarrier (about 3.58 MHz in NTSC) sitting inside the luminance band, itself a small-scale frequency-division multiplex. Telemetry and instrumentation systems have long used FM/FM schemes in which many sensor channels each frequency-modulate their own subcarrier and the whole stack then modulates the RF carrier. And in the digital world the idea generalizes completely: OFDM splits a channel into hundreds or thousands of closely spaced, individually modulated subcarriers, the technique behind Wi-Fi, LTE, 5G NR, DAB, and DVB-T. The humble FM pilot-and-subcarrier scheme and a 5G resource grid are points on the same continuum.
Relevance to SDR
Subcarriers are where a lot of “hidden” data lives in an otherwise analog broadcast, and they are a favorite SDR target. A software receiver that FM-demodulates a broadcast station gets the whole composite baseband for free and can then dig out the 57 kHz RDS subcarrier to display station name, program type, and RadioText — a common demonstration project. Decoding a suppressed-carrier subcarrier like the 38 kHz stereo channel requires regenerating it from the pilot, exactly the coherent-detection problem that double-sideband suppressed-carrier signals pose. GopherTrunk focuses on digital land-mobile trunking rather than broadcast, so it does not decode FM subcarriers, but the frequency-division-multiplex idea recurs throughout RF, and the pilot-regeneration technique is the same math its digital carriers use for synchronization.
Reading a subcarrier well depends on cleanly recovering the whole composite baseband first, which in turn needs enough receiver bandwidth and a low-distortion FM demodulator — a subcarrier riding at 57 kHz is useless if the receiver’s channel filter rolls off at 20 kHz. This is why subcarrier decoding rewards a wideband front end: the SDR must pass the full multiplex, not just the audio a listener hears. The same requirement applies in reverse as a warning about aliasing — a subcarrier sits at a real baseband frequency and must be sampled fast enough to represent it, or it folds back on top of the wanted audio.
Sources
-
Subcarrier — Wikipedia, for the definition and the FM stereo/RDS multiplex example. ↩