How is RF power measured accurately?
RF power is measured with power meters and sensors. Thermal sensors (thermistors, thermocouples) respond to true average power independent of waveform; diode sensors are faster and more sensitive, operating in square law at low levels (roughly below −20 dBm) and as peak detectors at higher levels. Accuracy depends on sensor calibration factor, linearity, zeroing and especially mismatch uncertainty.[1]
How large is mismatch uncertainty?
The limits are ±20·log10(1 ± |Γsource||Γsensor|) dB, often expressed as ≈ ±2|Γs||Γl| in linear power ratio terms. A source with VSWR 1.5 (|Γ| = 0.2) and a sensor with VSWR 1.2 (|Γ| = 0.09) gives about ±0.16 dB. Attenuator pads and good-match sensors reduce it.[1]
How are modulated and pulsed signals measured?
Average-power sensors work for any waveform within their range if diode sensors stay in square law or use multi-path designs. Peak and burst measurements need sensors with video bandwidth wider than the modulation and show envelope statistics such as CCDF and pulse parameters.[1]
What best practices apply?
- Zero the sensor before low-level measurements.
- Enter the correct calibration factor for the frequency.
- Allow warm-up; avoid temperature changes.
- Do not exceed damage level; use attenuators with known loss for high power.
Frequently asked questions
Is a spectrum analyzer as accurate as a power meter?
Typically less accurate in absolute power; power meters are the reference.
What is calibration factor?
Sensor efficiency and mismatch correction versus frequency, supplied with the sensor.
Why use thermal sensors?
They read true average power regardless of waveform.
See also
References
Online sources accessed October 21, 2026; printed works cited by edition. See all sources.
- “Fundamentals of RF and Microwave Power Measurements (Application Note 1449 series)”. Keysight Technologies. application note.