NLoS & Multipath Planning

Plan NLoS links before you deploy Tarana
(or any NLoS-capable radio)

GridVisio's LoS Link Check searches for one possible reflected signal path — off a nearby rooftop, water, or open ground — using the same real LiDAR terrain precision as its standard Fresnel check. It's a desktop planning aid that gives you an extra, honest data point before a site survey, not a model of the multipath signal reconstruction NLoS radios do internally.

Start 14-day free trial See the LoS Link Check tool

What this page is — and isn't

GridVisio is coverage-planning software, not an NLoS radio vendor — this isn't a "GridVisio vs. Tarana" comparison, because the two aren't the same kind of product. Tarana (and similar NLoS-capable platforms) reconstruct a usable signal from real multipath reflections in hardware, in real time, in ways no desktop terrain model can predict. GridVisio's NLoS reflection check does something much narrower: it tests one candidate reflection point geometrically against real terrain and reports whether that specific path looks clear. Useful extra context for planning around a Borderline or Obstructed direct link — not a substitute for your radio vendor's own link-budget tools or a real site survey.

GridVisio NLoS reflection path check on a real Obstructed link — the reflected path drawn on the map alongside the direct line, reflection offset, added path loss, combined RSSI estimate, and local clutter margin

A found reflection path, drawn on the map alongside the direct line — real LiDAR terrain, real coverage heatmap

Why this matters for NLoS deployments

A geometrically obstructed direct path doesn't automatically mean "no link" for NLoS-capable hardware — but knowing whether a plausible reflected path exists at all is useful before you commit a truck roll or a radio pair to a site.

One real candidate reflection, tested honestly

The search tries a handful of fixed lateral offsets on either side of the direct line and only accepts a candidate when both legs — transmitter to reflection point, and reflection point to receiver — are independently confirmed clear against real terrain. No candidate found, no claim made.

Real LiDAR, not a guess

Where free public LiDAR coverage exists — USGS 3DEP in the US, and eight other regional government sources (England, Wales, Scotland, Canada, France, Netherlands, Spain, New Zealand) — the reflection search uses real measured surface height, not a land-cover category estimate.

A number, not just a maybe

When a candidate reflection is found, GridVisio reports its lateral offset, the extra path loss it adds, and a combined RSSI estimate next to the direct-path-only number — power-summed, not phase-accurate, and clearly labeled as such.

Honest about the gap that's left

Real constructive/destructive interference depends on continuously-varying, sub-wavelength path-length differences that no terrain dataset can predict at any resolution. That's exactly the gap NLoS hardware's own real-time signal processing is built to close — not something GridVisio claims to model.

Dense suburban clutter, covered a different way

The single-reflector search above is the right tool when a real reflecting surface exists — a lake, a rooftop, an open field. Dense suburban and urban obstruction is usually a different problem: too many small, unresolvable scatterers (house facades, fences, tree canopy) to trace individually. For that case, GridVisio separately reports an ITU-R P.2108 statistical clutter-loss margin for whichever end sits below its local clutter height — no extra terrain fetch, no guessed land-cover category, just frequency, distance, and which end(s) qualify.

See the obstruction in 3D

A geometric verdict and an offset number only go so far when you're trying to picture what's actually in the way. GridVisio's 3D Corridor Check (Pro plan) renders the real LiDAR terrain and surrounding buildings along the link as an explorable 3D scene, with the same Fresnel geometry drawn as rings you can orbit around — useful for judging a candidate site before deciding whether it's worth a truck roll or an NLoS radio pair.

Where GridVisio fits — and where it stops

Three distinct levels of "does this link work," each doing a genuinely different job. GridVisio covers the first two; the third is what NLoS-capable radio hardware exists for.

Level	What it checks	Who does it
Tier A — Fresnel clearance	Is the direct line geometrically clear, using real terrain, land-cover, and LiDAR height? Clear / Borderline / Obstructed, with an exact clearance percentage.	GridVisio's standard LoS Link Check, every plan
Tier B — reflection path	Does one specific candidate reflected path exist and look clear on both legs? Reflection offset, added loss, and a combined RSSI estimate — power-summed, single-candidate, geometry only.	GridVisio's NLoS reflection path check (Pro/Trial)
Tier B+ — clutter margin	For a terminal embedded in dense clutter (too many small scatterers to trace individually), what's the statistically expected added loss (ITU-R P.2108) — a typical figure and a separate worst-case RSSI, not another discrete path.	GridVisio's local clutter margin, reported alongside the reflection check (Pro/Trial)
Real multipath reconstruction	Continuously adapting to real, time-varying multipath in the actual RF environment — the thing that makes an NLoS link genuinely usable in practice, not just theoretically plausible.	The NLoS radio hardware itself (e.g. Tarana), plus a real site survey

Frequently asked questions

Does GridVisio compete with Tarana?

No. Tarana makes NLoS-capable radio hardware; GridVisio is coverage-planning software. Many WISPs use both together — GridVisio for desktop planning, site inventory, and BEAD/BDC documentation across the whole network, and Tarana (or similar hardware) for the actual NLoS link. This page exists to explain honestly what GridVisio's reflection-path check does and doesn't do for that kind of deployment, not to suggest it replaces the radio's own engineering.

What exactly does the NLoS reflection path check compute?

On a Borderline or Obstructed direct link, an on-demand toggle searches a handful of fixed lateral offsets (both sides of the direct line) for one candidate reflection point using an image-method construction, then checks whether both legs of that specific path are clear against real terrain — the same real LiDAR data used elsewhere in GridVisio where it's available. If a candidate passes, GridVisio reports its offset, the added path loss, and a combined RSSI estimate (direct + reflected paths summed as power, not phase). If nothing passes, it says so plainly.

Is this the same as multipath modeling or ray tracing?

No, and GridVisio doesn't claim it is. Full ray tracing would search every possible reflection geometry and combine them with phase-accurate precision; this checks one plausible candidate per side and combines power only. Real-world constructive/destructive interference depends on continuously-varying, sub-wavelength conditions no static terrain dataset can capture — that's a genuinely different, harder problem than geometric obstruction detection, and it's what real hardware measurement and NLoS radio engineering exist to solve.

What about dense suburban links where there's no single obvious reflector?

That's exactly the case the reflection search isn't built for — it needs one identifiable flat surface, which dense clutter usually doesn't have. Alongside the reflection check, GridVisio separately reports an ITU-R P.2108 statistical clutter-loss margin for whichever end of the link sits below its local representative clutter height: a typical added-loss figure and a separate worst-case RSSI, based only on frequency, distance, and which end(s) qualify — no land-cover guess, no new terrain fetch. It's a different physical claim from the reflection check (aggregate statistical variability from many small unresolvable scatterers, not one traced path), so it's reported as its own number rather than folded into the combined RSSI above.

Should I still do a site survey if GridVisio finds a reflected path?

Yes. Treat a found reflection as a reason to take a closer look at a candidate site — not as a green light on its own. A real site survey and your radio vendor's own planning tools remain the authoritative step before any NLoS deployment.

Does this feature cost extra?

No — it's included wherever predicted RSSI already is, on the Pro plan and during your 14-day trial. It runs on demand only (a real extra terrain lookup), never automatically.

Plan NLoS links before you deploy Tarana(or any NLoS-capable radio)