Work out how many electronic shelf label gateways a store, warehouse or distribution centre needs. The tool corrects for mounting height — a base station fixed high above the shelf plane covers a smaller floor footprint than its rated slant range suggests.
Enter the space and link budget. Results update live.
Four inputs, one piece of geometry, and two multipliers. Nothing here is proprietary — you can reproduce it on paper.
Gateway datasheets quote a range as a straight-line distance through the air. But labels do not sit at ceiling height — they sit on a shelf edge, typically a metre or two off the floor. The distance that matters for planning is the horizontal one, and it is always shorter than the rated figure.
Treat it as a right triangle. The rated range is the hypotenuse, the height difference between gateway and labels is one leg, and the usable floor radius is the other:
In a supermarket the penalty is small: a 25 m gateway mounted 6 m up, with labels at 1.5 m, gives h = 4.5 m and r = 24.6 m — barely 2% lost. In a high-bay warehouse it is not small. Mount that same gateway on a 20 m ceiling and h becomes 18.5 m, leaving r = 16.8 m — a third of the reach, and 55% of the covered area, gone. Mount it high enough that h exceeds d and the coverage circle collapses to nothing, which is exactly what the tool warns about when it returns a zero radius.
| Layout | Spacing | Trade-off |
|---|---|---|
| Square grid | r × √2 between stations | Needs a few more units, but the positions land on ceiling grids, lighting rows and existing cable trays. Faster to install, easier to document, easier for a third-party contractor to get right. |
| Triangular grid | √3 r between columns, 1.5 r between rows | The efficient circle packing in the abstract, but the benefit only appears at scale. Offset rows need an extra station at each end to close the edges, so on small or near-square floors it costs more hardware, not less. |
Triangular placement is more area-efficient on an infinite plane, but a real floor has edges, cable routes and fixed mounting points. Neither layout is automatically smaller. The planner keeps the same radio assumptions for both layouts so you can compare the finite-floor effect directly.
The counts below use this page’s default coverage inputs: a 25 m rated range, a 6 m mount height and 1.5 m label height, giving a 24.6 m floor radius. They are coverage-planning counts only, before the UWB positioning uplift and before an RF site survey.
| Floor size | Square grid | Triangular grid | Difference |
|---|---|---|---|
| 2,000 × 500 m distribution centre | 870 | 665 | Triangular −24% |
| 500 × 300 m high-bay site | 135 | 112 | Triangular −17% |
| 200 × 120 m large-format store | 24 | 22 | Triangular −8% |
| 120 × 60 m supermarket | 8 | 7 | Triangular −13% |
| 100 × 100 m square floor | 9 | 10 | Triangular +11% |
| 60 × 40 m small store | 4 | 5 | Triangular +25% |
The practical rule: start with the square grid on a retail floor, where positions can align with ceiling grids, lighting rows and cable trays. Compare both layouts for a large, open footprint, then validate the apparent saving against power, network and RF constraints before ordering hardware.
These are different problems and they need different densities. Coverage asks a single question: can at least one gateway reach this label? Positioning asks something much harder: can several gateways measure this label at the same moment, so a location can be trilaterated from the overlap?
Selecting UWB positioning applies a 1.6× planning allowance on top of the coverage count. That is a planning heuristic, not a guarantee of a given accuracy figure — real positioning performance depends on anchor geometry, multipath in the aisle and the specific radio stack. Use it to budget, then validate on site. If indoor location is the actual goal, our note on retail indoor positioning covers what the technology can and cannot do.
One input dominates everything else: the comms range. Because coverage scales with the square of the radius, halving your assumed range roughly quadruples the station count. If your figure comes from an open-field datasheet number rather than an environment like yours, the result will be optimistic by a wide margin.
The safest approach is to model twice — once with a range you are confident about, once with a pessimistic one — and treat the gap as your procurement risk. Then confirm with a site survey before ordering. Gateway placement in a live store is also constrained by power availability and network drops, which no geometric model can see.
This planner produces a geometric baseline for budgeting and layout discussion. It is not a substitute for an RF site survey. For a validated design of your own site — including racking, power and network constraints — ask for a deployment assessment.
Model payback period and three-year net savings, with every assumption published.
Open the calculator →Why high-bay and cold-chain sites plan gateways differently from retail floors.
Read the guide →What has to be in place before the first gateway comes online.
See integration paths →The AES-B001 base station and the label range it drives, from peg hooks to pallets.
Browse products →Send us a floor plan and we'll return a placement layout with power and network requirements marked.
Talk to an engineerIt depends on floor area and effective coverage radius, not on label count. Divide the floor by the area each gateway can serve: with a square grid, one station covers roughly 2r², where r is the coverage radius measured at the label plane. A 120 × 60 m space with a 24 m effective radius needs about 6–8 stations for coverage. Very high label density or metal racking can require more.
A rated range is a slant distance measured through the air in a straight line, but labels sit on a lower plane. The usable floor radius is the horizontal leg of a right triangle: r = √(d² − h²), where d is the rated range and h is the height difference between the gateway and the labels. Mounting a 25 m gateway 4.5 m above the shelf plane leaves about 24.6 m of floor radius; mount it 18.5 m above the labels and only about 16.8 m remain.
It depends on the size and shape of the floor. Triangular placement is more area-efficient on an infinite plane, but finite-floor edges and fixed mounting points can reverse the count. With this planner’s default radio assumptions, a 2,000 × 500 m distribution centre uses about 24% fewer triangular-grid stations, while a 60 × 40 m store uses about 25% more. Start with square on a typical retail floor, compare both on a large open footprint, and validate the result with an RF survey.
Coverage only asks whether one gateway can reach a label. Positioning asks that several gateways see the same label at once, because a location is computed by trilateration from multiple simultaneous measurements. This planner applies a 1.6× density uplift as a planning allowance for that overlap requirement.
No. It models free-space geometry only. Dense metal racking, cold rooms, fire doors, lift shafts and reinforced concrete all attenuate signal and can meaningfully reduce the effective radius. Treat the result as a planning baseline and confirm it with an on-site survey before ordering hardware.
Use the range your gateway is rated for in an environment resembling yours, not the best-case open-field figure on a datasheet. If you are unsure, model a range you are confident about and a pessimistic one, then plan for the gap. Halving the assumed radius roughly quadruples the number of stations, so this input matters more than any other.
We'll return a gateway placement layout for your site, with power and network drops marked, plus the label mix each zone needs.
Request a deployment assessment