E8A: Fourier Analysis and Power Measurements

Understanding how signals are composed and measured underpins everything else in radio engineering. Fourier analysis reveals the frequency content hidden inside any waveform, analog-to-digital conversion bridges the analog and digital worlds, and proper power measurement ensures you know what your transmitter is actually doing.

This lesson covers Fourier analysis and waveform composition, the time and frequency domains, analog-to-digital conversion techniques (including dither and flash converters), true-RMS measurements, peak envelope power versus average power, and quality metrics for ADCs.

Fourier Analysis and Signal Composition

Fourier analysis is a mathematical technique that decomposes any periodic waveform into a sum of sine waves at different frequencies. The central insight is that complex waveforms are built from simple sinusoidal components.

A classic example is the square wave: Fourier analysis shows that a square wave is composed of a fundamental sine wave and all of its odd harmonics (3rd, 5th, 7th, and so on), each with decreasing amplitude. This is why a square wave transmitted on an amateur radio transmitter produces harmonic energy at odd multiples of the fundamental frequency — and why filtering is required to prevent interference.

Time Domain vs Frequency Domain

Signals can be viewed in two complementary domains:

• Time domain: Describes a signal as amplitude at different times — the familiar oscilloscope view, where you see how the signal voltage varies moment to moment.
• Frequency domain: Describes a signal as amplitude (or power) at different frequencies — the spectrum analyzer view, where you see which frequency components are present.

Fourier analysis mathematically converts between these two representations. Understanding which domain you are working in is fundamental to signal analysis.

Analog-to-Digital Conversion

An analog-to-digital converter (ADC) samples a continuous analog signal and converts each sample to a digital number. The number of distinct levels an ADC can represent depends on its bit resolution:

One common type of ADC is successive approximation, which works by testing one bit at a time from the most significant to the least significant, converging on the correct digital value. It is efficient for moderate speed applications.

Dither and Quantization

Quantization is the process of rounding an analog sample to the nearest digital level. When the input signal is very small or slowly varying, the ADC may repeatedly produce the same output code, creating a distortion called quantization noise.

Dither is a small amount of noise intentionally added to the input signal before sampling. By randomizing which quantization level is selected, dither converts the structured quantization distortion into low-level broadband noise — which is far less objectionable and easier to filter. Dither is not an error; it is a deliberate technique for improving perceived ADC performance.

Flash Conversion for SDR

A flash (direct) conversion ADC uses a bank of comparators — one for each possible output level — to convert the entire input voltage in a single clock cycle. This architecture achieves extremely high conversion speeds, making it suitable for digitizing RF signals directly at high frequencies in software defined radios (SDRs). The trade-off is high power consumption and large die area, but for SDR applications, the speed advantage is decisive.

True-RMS Measurements

Root mean square (RMS) voltage is the equivalent DC voltage that would produce the same heating effect in a resistor. A standard AC voltmeter assumes the signal is a pure sine wave and applies a fixed conversion factor. This gives correct results for sine waves but produces errors for non-sinusoidal waveforms such as square waves, audio signals, or RF envelopes.

A true-RMS calculating meter mathematically computes the actual RMS value regardless of waveform shape. Its benefit is that it measures RMS correctly for both sinusoidal and non-sinusoidal signals — making it essential for accurate power measurements on complex or modulated waveforms.

PEP and Average Power for SSB

In single-sideband (SSB) phone operation, two power values matter:

• Peak envelope power (PEP): The power at the peak of the modulation envelope — the highest instantaneous power during a syllable.
• Average power: The power averaged over a period of time, which varies with the audio content.

For an unprocessed SSB phone signal, the PEP-to-average power ratio is approximately 2.5 to 1. This ratio is determined by speech characteristics — the natural variation in loudness and silence during conversation. When speech processing (compression) is applied, the ratio decreases as average power is raised closer to PEP.

DAC Low-Pass Filtering

After a digital-to-analog converter (DAC) reconstructs a waveform from digital samples, the output contains unwanted high-frequency artifacts — staircase steps and spectral aliases — that arise from the sampling process. A low-pass filter at the DAC output removes these spurious sampling artifacts, leaving only the desired reconstructed analog waveform. This is sometimes called a reconstruction filter.

ADC Quality Metrics

The quality of an ADC is most commonly measured by its total harmonic distortion (THD). THD quantifies how much harmonic distortion the converter introduces when digitizing a pure sine wave input. A lower THD indicates a higher-fidelity converter with less nonlinear distortion — critical for applications such as SDR receivers where the ADC directly digitizes RF signals.

E8A Practice Questions