Also known as: spectrum analyser, SA, RTSA
A spectrum analyzer is an instrument that displays signal power as a function of frequency1 — the frequency-domain counterpart to an oscilloscope’s time-domain trace. It answers “how much energy is present at each frequency?”, making it the primary tool for finding carriers, measuring occupied bandwidth, hunting spurious emissions, and characterizing the phase noise of an oscillator.
Overview
Two things define what a spectrum analyzer can show you: its noise floor (how weak a signal it can resolve) and its resolution bandwidth (RBW, how finely it separates two close-spaced tones). A narrower RBW lowers the displayed noise floor and sharpens adjacent signals, but each halving of RBW roughly quadruples sweep time on a swept instrument. Span, reference level, and detector mode round out the core controls.
Swept vs. FFT / real-time
Classic bench analyzers are swept-tuned superheterodyne receivers: a local oscillator ramps across the span, mixing each frequency in turn down to a fixed IF where a narrow filter and a detector measure power. Because the LO visits one frequency at a time, a swept analyzer is blind between visits — a short burst that occurs while the sweep is elsewhere is simply missed. Swept designs excel at wide spans with excellent dynamic range and low noise, which is why high-end phase-noise and spurious measurements still favor them.
FFT analyzers instead digitize a block of the input and compute a fast Fourier transform, producing all frequency bins in that block at once. This is exactly how a software-defined radio or the tinySA in its lower bands measures spectrum. FFT processing is fast over a limited span and yields the power spectral density directly, but the instantaneous span is capped by the ADC sample rate.
A real-time spectrum analyzer (RTSA) overlaps successive FFTs so that no input gap exists between transforms — it guarantees 100% probability of intercept for any event longer than a stated minimum duration. RTSAs add persistence and density displays (a spectrogram or bitmap-style overlay) that reveal transient and intermittent signals a swept trace would average away. Modern instruments frequently combine a swept front end for wide spans with an FFT/real-time block for detailed, gap-free analysis over a narrower window.
In practice
Key settings to reason about on any analyzer:
- RBW and VBW — resolution bandwidth sets frequency selectivity and noise floor; video bandwidth smooths the trace.
- Detector and trace mode — peak, sample, RMS/average, and max-hold change what a bin reports; RMS is correct for noise-like signals, max-hold catches transients.
- Reference level and attenuation — set the top of the screen and protect the front end; too little attenuation invites internally generated intermodulation that masquerades as real spurs.
- Dynamic range — the window between the noise floor and the onset of front-end compression or spurious products bounds what you can measure at once.
Relevance to SDR
A spectrum analyzer is the natural companion to SDR scanning: use it to survey a band, confirm which control-channel and voice frequencies are actually active, gauge signal strength, and diagnose interference, images, and de-sense before committing an SDR to decode. Any SDR receiver is a rudimentary FFT spectrum analyzer — GopherTrunk’s own waterfall and FFT views plot power versus frequency the same way — though a dedicated instrument offers calibrated amplitude, far better dynamic range, and a much lower noise floor. GopherTrunk does not drive external analyzers; it uses its internal FFT for tuning and diagnostics, and a bench or handheld analyzer remains a useful external aid for setting up antennas and filters.
Sources
-
Spectrum analyzer — Wikipedia, on swept-tuned, FFT, and real-time spectrum analyzer architectures and their measurements. ↩