Does choosing UHF over VHF actually fix basement and stairwell dead zones?

No. UHF is the right band for an in-building system in Singapore, but the advantage it gives you over VHF in reinforced concrete is on the order of 5 to 10 dB. A basement carpark two levels below grade typically sits 30 to 50 dB below the ground-floor lobby level. Band choice moves the number in the right direction and does not come close to closing the gap. If your radios do not work in the basement on VHF, they will not work in the basement on UHF either - you need a repeater and distributed antennas, and the band is a separate decision you make before you buy the hardware.

I keep seeing "choose UHF" offered as the fix for below-grade dead zones. It is the correct band and the wrong answer to the question. Below is the arithmetic that settles it, and the part of the UHF-versus-VHF story that most explanations get backwards.

Why UHF is better in concrete, when it actually has higher loss

The common claim is that UHF "penetrates" better than VHF. Taken literally that is false, and the sources that repeat it often contradict themselves in the same paragraph. Through a single plasterboard partition, 150 MHz loses roughly 3 dB and 450 MHz roughly 6 dB - the higher frequency loses more. Free-space loss is higher too: 20 log(450/150) = 9.5 dB more at the same distance. On raw attenuation, VHF wins.

UHF still wins indoors, for three reasons that have nothing to do with attenuation through a solid slab:

• Rebar acts as a screen, and screens leak more at higher frequencies. A typical reinforced-concrete mat has bar spacing of 150 to 200 mm. At VHF the wavelength is 2 m, so the mesh apertures are roughly a thirteenth of a wavelength and the mat behaves close to a solid conductive sheet. At 450 MHz the wavelength is 667 mm and the same apertures are about a quarter of it, so the mat is a far less effective shield. This is where the real in-building advantage comes from, and it is why the gap is largest in exactly the heavy RC construction Singapore builds in.

• Openings. Signal reaches a stairwell or a carpark through door gaps, riser penetrations, ventilation openings and lift-door clearances - apertures of 50 mm to 1 m. A 667 mm wavelength couples into those; a 2 m wavelength largely does not.

• Antenna efficiency on the handset. A quarter-wave at 450 MHz is about 165 mm, which is a real, efficient radiator on a portable. At 150 MHz a quarter-wave is 500 mm, so VHF portables run heavily loaded stubby helicals that give up several dB of efficiency. That handset penalty typically eats the propagation advantage on its own.

Net of all of it, in reinforced-concrete construction, expect UHF to deliver something in the region of 5 to 10 dB better than VHF at the same point in the building. That is a real margin and it is worth having. It is not a coverage solution.

The arithmetic that decides it

Take the standing engineering figures we use across our designs: a 200 mm reinforced-concrete floor slab costs 15 to 25 dB; an RC core or shaft wall costs 25 to 35 dB; reliable DMR voice needs -95 dBm at the portable, with 10 dB of fade margin for a commercial system and 15 dB where it is mission-critical.

Suppose you measure -70 dBm in the ground-floor lobby, which is a healthy reading. Drive down two levels to B2 and you are through two slabs: 30 to 50 dB of additional loss, putting you at -100 to -120 dBm. You are 5 to 25 dB below threshold before any fade margin. Add a 5 to 10 dB band improvement and you are still below threshold in almost every case, and you have spent your entire hardware budget to get there.

There is a second problem band choice does nothing about. A 40 W repeater transmits at +46 dBm; a 4 W portable transmits at +36 dBm and loses about 3 dB more to the operator's body. That is roughly 13 dB of built-in asymmetry in favour of the downlink. It is why a security officer in a carpark often hears the guardhouse faintly and cannot be heard back - and why a coverage survey that only measures downlink signal will pass a system that fails in service. Switching bands leaves that 13 dB exactly where it was.

So when does the band decision matter?

It matters at procurement, and it is expensive to reverse. VHF and UHF are different radios, different antennas and different feeder hardware - re-programming will not move a fleet between them - and IMDA assigns private land mobile spectrum band by band, with VHF at 137-174 MHz and UHF at 400-450 MHz. Choose UHF at the start for any building with concrete floor plates, and treat that as the baseline, not the remedy.

The remedy is a repeater with a distributed antenna system that puts a radiating point inside each below-grade level, or radiating cable where the space is long and narrow, such as a drive aisle or a ramp. Design it to -95 dBm with 10 dB of margin, or 15 dB in stairwells and the fire command centre, and verify it with a grid survey measured in both directions. Singapore has no IFC 510-style statutory coverage code; the -95 dBm figure and the 95% area target are contractual, so they need to be written into the tender or nobody owes you them.

Related questions

Would switching from analogue to digital DMR help instead? No, for the same reason. DMR gives you roughly 2 to 3 dB of usable sensitivity improvement from error correction, not the tens of dB a basement is short by. It also fails abruptly rather than gracefully at the threshold.

Does a higher-gain antenna on the handset close the gap? Marginally. A full-length whip in place of a stubby is worth perhaps 2 to 3 dB, and it is the cheapest few dB available - but it is still single-digit dB against a 30 to 50 dB deficit.

Is 800 MHz better still, since higher is better in concrete? No. The rebar-screen argument runs out; above roughly 800 MHz the free-space and material losses dominate and coverage per antenna falls sharply. UHF at 400-470 MHz is the practical optimum for in-building two-way radio.

Suneast designs and installs in-building two-way radio coverage systems in Singapore - repeaters, bi-directional amplifiers and distributed antenna systems for carparks, stairwells, lift shafts and plant areas. We survey the building first, compute the link budget in both directions, and hand over a grid-tested system against a stated dBm threshold rather than a promise about the band. Related reading: why radios fail in basement carparks and stairwells and how much signal a radio loses through each concrete slab.