Why doesn't a walkie-talkie's advertised 300,000 sq ft coverage rating translate into floors covered in a concrete building?

Because the rating is a horizontal, single-floor, open-space figure, and floors are not a distance problem — they are an attenuation problem. A 300,000 sq ft claim describes roughly a 94 m radius across one unobstructed floor plate with no slab in the signal path. Each 200 mm reinforced-concrete floor slab costs 15–25 dB, and two of them can consume more of the link budget than the entire advertised footprint does. On a realistic budget, the same handset that is sold as covering "25 floors" will punch through two to three reinforced-concrete slabs before it drops below a usable level.

What does a 300,000 sq ft coverage rating actually describe?

It describes area, not volume, and it is measured under conditions that flatter the radio. Convert it and the claim becomes legible: 300,000 sq ft is about 27,900 m², which as a circle is a radius of roughly 94 m. That is a single open floor plate — a warehouse, a retail hall, a car park deck — with the two radios at similar height and nothing structural between them.

Nothing in that figure describes vertical penetration. When a datasheet pairs it with "up to 25 floors" or "up to 30 floors", the two numbers are not derived from the same test. The floor count is the manufacturer's separate best-case observation, usually in a lightly built structure, and it is not an engineering limit you can design a building against.

How many floors does that budget really buy in reinforced concrete?

Run the link budget rather than trusting the box. Take a 4 W portable at 450 MHz:

• Portable transmit power: 4 W = 36 dBm; body loss at the hip: 3 dB; effective radiated level ≈ 33 dBm

• Design threshold for reliable DMR voice: −95 dBm (not the −119 dBm receiver sensitivity)

• Fade margin: 10 dB standard commercial, 15 dB for mission-critical or public-safety areas

• Allowable path loss after margin: about 118 dB

• Indoor path-loss exponent: 3.0–3.5 (3.2 assumed here)

• 200 mm reinforced-concrete floor slab: 15–25 dB; reinforced-concrete core and shaft walls: 25–35 dB

Now the two cases. Across one open floor at 94 m, path loss is roughly 25.5 + 32·log₁₀(94) ≈ 89 dB — comfortably inside the 118 dB allowance, which is why the advertised area figure is achievable in the environment it was measured in. Vertically, take a portable 20 m horizontally and four floors away, a slant distance of about 25 m: path loss is only around 70 dB, leaving about 48 dB. At 20 dB per slab, that is 2.4 slabs. Two to three floors of reinforced concrete, against a claim of 25.

The distance term is not what limits you in a tower. The slabs are. This is the same arithmetic behind our figures on how much signal a radio loses through each concrete floor slab and basement wall.

Why does the floor count on the box scale so badly?

Because the marketing arithmetic treats the area figure as something you can slice across storeys — 300,000 sq ft divided by 25 floors is 12,000 sq ft per floor, which sounds reasonable and is meaningless. It assumes each slab is free. In a Singapore reinforced-concrete tower with post-tensioned slabs, dense rebar mesh and a central RC core carrying the lifts and stairs, the slab is the dominant loss term in the entire link.

There is a second reason the rating misleads, and it is the one that catches sites out after purchase. Advertised figures are radio-to-radio talk-around numbers under favourable conditions. They say nothing about the uplink into a fixed repeater, where a hip-worn 4 W portable behind three slabs has to reach a rooftop receiver. That path fails first, and it fails quietly — staff hear the repeater and assume the system works.

Does buying a higher-power radio fix it?

Only marginally, and not in the way buyers expect. Going from 4 W to 5 W is 1 dB. Against a 20 dB slab, 1 dB buys nothing. Even doubling power is 3 dB, roughly one-seventh of one floor. Power is the wrong end of the link to solve a penetration problem, which we set out in detail in whether more repeater power or a higher-gain antenna fixes coverage. The fix for vertical loss is to stop trying to pass through the slabs — put antennas on the floors that need coverage and feed them from a riser.

How should a specification be written instead?

Never in square feet. Specify a measured signal level and an audio quality grade over a defined percentage of floor area — for example −95 dBm or better with DAQ 3.0 across 95% of general areas and 99% of critical areas such as stairwells, lift lobbies, basement car parks and the fire command centre, verified by a grid survey in both directions. A number written that way can be tested at handover. A square-footage rating cannot be tested at all, which is precisely why it appears on packaging and never in a tender.

Related questions

Is the 300,000 sq ft figure simply false? No — it is a real measurement under stated conditions, and roughly a 94 m radius on an open floor plate is achievable. It is misapplied, not fabricated, and the error is treating a horizontal open-space figure as a multi-storey design rule.

How many floors can one portable realistically reach in a Singapore RC tower? On the budget above, two to three slabs from the transmitting radio, less if the path crosses the RC core or a lift shaft. Anything beyond that needs a repeater with distributed antennas.

Does a 900 MHz digital handset quoting 350,000 sq ft do better? No — higher frequencies attenuate more through concrete, not less. A larger advertised area at a higher band usually reflects a different test environment, not better penetration.

How Suneast approaches this

We do not quote coverage in square feet. For every in-building job we run a link budget for both downlink and uplink, state the assumptions and defaults used, and produce an antenna placement plan and coupler tap schedule that a contractor can build from and a consultant can recompute. Coverage is then verified on site by grid survey against a stated dBm threshold and audio quality grade, in both directions, with the results handed over as a document. If a design is marginal, we say so and show the margin rather than quoting a footprint.