What frequency range does the Okumura-Hata model cover?

The Okumura-Hata model is the most widely used empirical model for macro-cell mobile radio planning in the 150–1500 MHz range. It curve-fits Okumura's original measurement data into closed-form expressions for urban, suburban, and open rural environments, with typical prediction errors of ±8–12 dB (one standard deviation) in the environments for which it was calibrated.

What is the difference between the Hata model and COST 231?

The Hata model covers 150–1500 MHz, while the COST 231 project extended it to the 1500–2000 MHz range (PCS and early 3G bands), producing the COST 231-Hata model. COST 231 adds a metropolitan area correction factor C_m of 0 dB for medium-sized cities and suburbs, or 3 dB for metropolitan centres, and is standard in GSM 1800 and UMTS 2100 planning. Beyond 2 GHz, accuracy degrades and ray-tracing or measurement-based models are preferred.

What is the Longley-Rice (ITM) propagation model used for?

The Irregular Terrain Model (ITM), developed at NTIA/ITS, is the primary propagation model used by the FCC and international regulators for point-to-point and broadcast services from 20 MHz to 20 GHz. Unlike Hata, it is physics-based and ingests actual terrain profiles, antenna heights, frequency, polarisation, and required reliability percentages, computing diffraction loss over terrain obstacles with a refined knife-edge method. The FCC uses ITM in its OET Bulletin 69 tools for FM and TV coverage prediction.

What is the difference between Rayleigh and Rician fading?

Rayleigh fading occurs when there is no dominant line-of-sight component, representing the worst case, with average fade depth at the 1% probability level approximately 20 dB below the median. Rician fading applies when a strong LOS component exists alongside scattered components, characterised by a K-factor (ratio of LOS power to scattered power) — typical K values are 0–3 dB for urban mobile, 6–12 dB for microcell LOS, and 15–25 dB for fixed point-to-point links with clear Fresnel zone clearance.

How much diffraction loss does a knife-edge obstacle cause?

Knife-edge diffraction loss is L_d ≈ 6.02 + 9.11ν + 1.27ν² dB for the Fresnel-Kirchhoff diffraction parameter ν between 0 and 2.4, where ν depends on obstacle height above the direct ray and the path geometry. To keep diffraction loss under 6 dB, at least 0.6 of the first Fresnel zone radius must remain unobstructed along the path.

Engineering·11 min read·June 17, 2026

📡 RF Propagation Models: From Free Space to Multipath Fading

A rigorous survey of RF propagation models used in network planning — Okumura-Hata, COST 231, Longley-Rice, indoor models, Rayleigh/Rician fading, log-normal shadowing, and practical frequency selection guidance.

📖Radio Communications — Illustrated Guide (Full Access)

View →

Free Space vs. Real-World Propagation

Free-space path loss (FSPL) describes signal attenuation in an unobstructed vacuum. Real-world environments impose additional losses from reflection, diffraction, scattering, absorption, and multipath interference. The excess path loss beyond FSPL is termed the propagation model correction factor, and it varies with frequency, antenna height, terrain, clutter, and building density. All empirical propagation models attempt to quantify this excess loss as a statistical function of measurable parameters.

Topics covered

RF propagationOkumura-HataCOST 231Longley-RiceITM modelRayleigh fadingRician fadinglog-normal shadowingknife-edge diffractionindoor propagationmultipathfrequency selection