Coverage prediction over real terrain

Provide a transmitter — location, antenna height, EIRP and frequency — and an area. The service returns a predicted signal-strength field computed with the Longley–Rice Irregular Terrain Model (ITM) over USGS 3DEP elevation data, with terrain diffraction, landcover loss and clutter applied along each path.

It was built as a faster alternative to the Radio Mobile workflow, which takes minutes per map. The same prediction here is a short wait.

How it is priced Pick a resolution that is worth paying for

What it computes

Single transmitter

One site, one frequency, one antenna height, out to a range you choose. Output is a signal-strength field, not a yes/no contour — you set the threshold afterwards.

Combined coverage

Several transmitters combined in two ways: per-pixel best server for trunked and roaming systems, or sum of power for simulcast and total illumination. They answer different questions and the difference is visible.

Antenna patterns

A built-in catalogue of named patterns — omnidirectional, cardioid, Yagi, corner reflector, and named commercial LMR models across low band, VHF, 220, UHF, 700, 800 and 900 MHz. Both the azimuth and the elevation pattern are applied, with the antenna pointed where you point it.

Time variability

The ITM Y_T term is applied, so "where do I have signal 50% of the time" and "99% of the time" are different maps and you can ask for either.

What it computes it over

Terrain. 1 m USGS 3DEP LiDAR where it has been flown; 1/3 arc-second (roughly 10 m) across CONUS everywhere else. The tiers are blended, not switched, so there is no step at the boundary.
Diffraction. Diffraction over terrain obstructions is calculated, rather than treating the path as free space with an empirical correction.
Clutter and ground cover. Landcover attenuation, building footprints as knife edges, and tree-canopy loss at the receiver.

What comes out

A georeferenced map overlay you can drop on a basemap.
A tile pyramid, for panning and zooming the result rather than re-rendering it.
KMZ export, for Google Earth and any other application that reads KMZ.

What actually drives the cost

The following factors are not obvious from the outside, and together they determine whether a render is worth its cost.

×4 per doubling of the grid.

Cost is based on the finished image, so it rises with the square of the grid size. Four doublings is 256×. This is the setting most often increased without appreciating its effect. Supersampling also multiplies cost by its square, and each additional transmitter adds a multiple; all three are in the formula.

10 m terrain posts across most of CONUS; 1 m where LiDAR has been flown.

Requesting a grid finer than the underlying terrain does not create detail. It interpolates, at four times the price per doubling.

268 MP in a mean of 152 s, observed over 29 renders at 16384².

This is the largest size we have measured. Processing time varies with demand on the service, so this figure is an observation, not part of what you purchase.

The most common sizing mistake

Both directions cost money. Requesting too fine a grid is the common and expensive mistake: if the terrain has 10 m posts, a finer grid returns no new information about it. Very small renders are also proportionally more expensive per megapixel, because every render is rounded up to a whole credit. Is my request too fine? works through how to size a grid so you are paying for detail that exists.

Who it is for

Repeater owners and trustees siting or re-siting a repeater.
Amateur and commercial VHF/UHF system operators planning simulcast or voting systems.
Public-safety and volunteer-emergency groups documenting the coverage they actually have.
RF engineers who want ITM over good terrain data without running it themselves.

What this is not

It is a propagation model, and every number it produces is a prediction with error bars that Longley–Rice itself does not hide. It is not a measurement, it does not replace a drive test, and it is not evidence of anything to a regulator: it performs no FCC coordination, produces no contour study anyone is obliged to accept, and makes no claim about interference to or from any licensed facility. It models outdoor path loss over terrain and ground cover. It does not predict in-building penetration or model your receiver's front end, and it will be wrong wherever the terrain data and statistical clutter are wrong.

Antennas are chosen from a built-in catalogue of named patterns. You cannot import your own .PAT or .NEC files, and sectorized panel antennas of the cellular and point-to-point kind are out of scope.

Treat its output as you would any ITM study: as an inexpensive first estimate before field testing, and as something field measurement takes precedence over.