What is RF design?
RF design is the engineering of circuits and systems that generate, amplify, filter, transmit, receive and process signals at radio frequencies, which the ITU defines as the spectrum below 3,000 GHz used for radio communication.[1] What separates it from low-frequency electronics is that wavelengths become comparable to circuit dimensions, so wiring behaves as transmission lines, impedance must be controlled and matched, and noise, linearity and radiated emissions dominate the design trade-offs.[2]
What frequency range counts as RF?
ITU-R Recommendation V.431 divides the radio spectrum into decade bands numbered 4 to 15, from VLF (3–30 kHz) to THF (300–3,000 GHz).[3] In everyday engineering usage, “RF” often means roughly 3 kHz to a few GHz and “microwave” roughly 300 MHz or 1 GHz to 300 GHz, with “millimeter wave” for 30–300 GHz (wavelengths of 10 mm down to 1 mm). The boundaries are conventions, not physics. See frequency bands.
Why is RF design different from ordinary circuit design?
A rule of thumb is that once an interconnect is longer than about one-tenth of a wavelength, its phase shift and characteristic impedance can no longer be ignored, and lumped-circuit analysis gives way to distributed analysis.[2] At 2.4 GHz the free-space wavelength is about 12.5 cm and on FR-4 microstrip roughly 7 cm, so a 7 mm trace is already a tenth of a wavelength. Other RF-specific concerns:
- Impedance matching to deliver power and avoid reflections (impedance matching).
- Noise: receivers are limited by thermal noise of about −174 dBm/Hz at 290 K (thermal noise).
- Linearity: compression and intermodulation set dynamic range (IP3).
- Parasitics: every capacitor has inductance, every via has inductance, every pad has capacitance.
- Radiation and coupling: unintended antennas cause EMI.
What are the main building blocks?
| Block | Function | Key metrics |
|---|---|---|
| LNA | First gain in a receiver | Noise figure, gain, IIP3 |
| Power amplifier | Drives the antenna | Output power, PAE, ACPR/EVM |
| Mixer | Frequency translation | Conversion gain, NF, IIP3, isolation |
| Filter | Select or reject frequencies | Insertion loss, rejection, group delay |
| Oscillator / PLL | Generate the LO | Phase noise, tuning range, spurs |
| Antenna | Couple to free space | Gain, efficiency, bandwidth, pattern |
What skills does an RF engineer need?
Electromagnetics and transmission-line theory, two-port network theory and S-parameters, the Smith chart, noise and nonlinear analysis, PCB layout, and laboratory measurement with a VNA and spectrum analyzer. Standard texts include Pozar for microwave engineering[2] and Razavi for RF integrated circuits.[4]
Frequently asked questions
Is RF the same as microwave engineering?
They overlap. Microwave usually refers to roughly 300 MHz–300 GHz where distributed design dominates; RF is often used for the whole radio spectrum.
What is the highest radio frequency?
The ITU Radio Regulations define radio waves as electromagnetic waves below 3,000 GHz propagating without artificial guide.
Do I need to know Maxwell's equations?
For circuit work, transmission-line and network theory suffice most of the time; antennas, EM simulation and EMC require field theory.
See also
References
Online sources accessed October 21, 2026; printed works cited by edition. See all sources.
- “Radio Regulations”. International Telecommunication Union (ITU-R). 2024 edition.
- “Microwave Engineering, 4th ed.”. D. M. Pozar, Wiley. 2012.
- “Recommendation ITU-R V.431: Nomenclature of the frequency and wavelength bands used in telecommunications”. ITU-R. current version.
- “RF Microelectronics, 2nd ed.”. B. Razavi, Prentice Hall. 2012.