How Can We Help?

Propagation models

You are here:
< All Topics

This service offers a variety of propagation models for different purposes:

Irregular Terrain Model (ITM)‘Longley Rice’ Model. US Gov. general purpose model used by FCC2-20,000MHz
Line of Sight (LOS)Simple model for viewing obstructionsAny
Okumura-HataHata model for cellular communications in urban areas150-1500MHz
ECC33 (ITU-R P.529)ECC33 model for cellular and microwave communications700-3500MHz
Stanford University Interim (SUI)SUI model for WIMAX communications1900-11000MHz
COST231-HataEuropean COST231 frequency extension to Hata model for urban areas150-2000MHz
ITU-R P.525Free Space Path Loss modelAny
Ericsson 9999Ericsson 9999 model for cellular communications up to 1900MHz150-1900MHz
Egli VHF/UHFGeneral purpose VHF/UHF model. More conservative than FSPL30-1000MHz


The SLEIPNIRTM engine can employ a given model beyond its stated limits (eg. 30m Tx height for Hata). This is handy when a model is required for a piece of spectrum which falls outside the stated limits of traditional models. Model specific frequency and height limits for some models are enforced on the CloudRF API for the sake of accuracy and to avoid errors.


This service uses the Irregular Terrain Model version 7, also known as the ‘Longley Rice’ model.
The ITM model is a long standing general purpose model developed by the US NTIA and used by the FCC which meets most radio engineering requirements for frequencies between 20 MHz and 20 GHz. It is suitable for everything RF from handheld VHF walkie-talkies to SHF microwave links. It factors in electromagnetic theory, terrain features, ground clutter, diffraction and radio measurements to predict the attenuation of a radio signal at a given point on the earth.

With the addition of reliability (time and situations) as a percentage the model can be tuned to be compliant with TSB-10 (Revision F) standards for microwave communications which defines a reliability of 80% for a long term loss and 99% for a short term loss.


The LOS model is a simpler model well suited to super high frequency microwave links where LOS is key. It factors in terrain data and antenna heights to provide a clear yes/no result. It starts where the ITM model finishes at 20GHz and extends the maximum range up to 100GHz.


The Hata models were designed for Urban cellular planning which is evident by their focused frequency range (150-1500MHz) and minimum tower elevation of 30m. The model assumes the transmitter is higher than the average height of the rooftops.


The COST231-Hata model extends the Hata cellular models another 500MHz up to 2000MHz so is well suited to GSM1800 in urban areas.


The free space model is a very simple model that assumes no obstacles exist between transmitter and receiver. It factors in the frequency and distance. It can be further enhanced with the addition of knife edge diffraction which will factor in terrain obstacles and antenna heights.

ITU-R P.529 ECC33

ECC33 model for cellular and microwave communications covering 700-3500MHz


General purpose VHF / UHF model for 30MHz to 1000MHz. It is more conservative than FSPL but more optimistic than Hata/COST.


The Irregular Terrain Model with obstructions 3.0 model is claimed by its author to be an enhancement to the ITM model with extra diffraction. Due to uncertainty over it’s Copyright (see SPLAT! changelog) it was removed from the SLEIPNIRTM engine.


In their native form, the empirical models (Hata, COST, ITU-R P.525) do not factor in the significant effects of diffraction caused behind obstacles. To accommodate this without the processing overhead associated with recognised complex diffraction theories, a proprietary knife edge diffraction function has been written which simulates modest diffraction, varied by wavelength and the angle of the received point to the obstacle summit. This can be enabled at the API with the ked parameter.


SLEIPNIRTM is the multi-core Cloud-RF propagation engine which was written from the ground up to work with high resolution LiDAR surface models and 3D clutter. It’s up to 10 times faster than open source engines for some scenarios and has unique capabilities such as multi-resolution modelling to create seamless surface models from patchy data.

Table of Contents