Install Quick Surf Network
Tap the Share icon
then Add to Home Screen to use QSN like an app.
Signal strength is the number everyone looks at and the number that explains the least. A room can show full bars and still be unusable, because what actually decides your experience is how far the signal sits above the noise around it. Here is how to read wireless properly — and what to do about what you find.
Everything else in this guide hangs off these two. Learn them and most wireless complaints become diagnosable.
RSSI stands for Received Signal Strength Indicator. Think of it as volume. It is written in dBm — decibels relative to one milliwatt — and it is always negative, because the power involved is a tiny fraction of a milliwatt. A reading of −45 dBm is loud and close. A reading of −80 dBm is faint and far. The counter-intuitive part is that −45 is stronger than −80, because it is closer to zero.
SNR stands for Signal-to-Noise Ratio. Think of it as clarity. If RSSI is how loudly someone is speaking, SNR is how loudly they are speaking compared with the noise in the room. Two people can shout equally loudly, but only one of them is in a quiet room. SNR is written as a plain number in dB and it is always positive: signal minus noise. A signal of −55 dBm sitting on a noise floor of −95 dBm gives an SNR of 40 dB, which is excellent. The same −55 dBm signal in a room where the noise floor is −65 dBm gives an SNR of 10 dB, which barely works.
This is why a phone showing four bars can still stream badly, and why replacing the router rarely helps. The bars are showing you volume. The problem is clarity. If wired versus wireless is the underlying question for your building, the wired vs wireless guide is worth reading alongside this one.
These are the numbers we design to. They are not universal law, but they are close enough to be useful in every building we have surveyed.
Applied to a real site: a published villa audit measured 14 rooms below this threshold — despite 32 access points being installed →
Three units that look alike, sound alike, and get mixed up in product listings constantly.
One asymmetry that catches people out: a powerful access point can be heard by a phone across the building, but the phone cannot shout back that far. Its antenna is smaller and its battery smaller still. Turning transmit power up on an access point often makes coverage look wider on a survey while making the actual experience worse, because devices connect from places they cannot hold a conversation from. This is one of the commonest self-inflicted wireless faults we are called to fix.
The single most common cause of a slow 2.4 GHz network, and it is free to fix.
This is not theoretical. In one audited estate, 15 of 30 radios sat on channel 6 — queuing for the same airtime instead of adding capacity →
2.4 GHz gives you three lanes. The other two bands are far more generous — but only if you choose the channel width honestly.
| Clean lanes available | 2.4 GHz | 5 GHz | 6 GHz |
|---|---|---|---|
| At 20 MHz | 3 | ~25 | 59 |
| At 40 MHz | 1 — do not | ~12 | 29 |
| At 80 MHz | Not possible | 6 (2 without DFS) | 14 |
| At 160 MHz | Not possible | 2 (DFS only) | 7 |
| DFS required? | No | For most of the band | No |
| Who can use it | Everything | WiFi 5 and newer | WiFi 6E and 7 only |
| Range | Furthest | Moderate | Shortest |
| Typical noise | Very high | Moderate | Very low |
Read the table as a planning tool rather than a scoreboard. If a building needs six access points, then at 80 MHz on 5 GHz you have exactly enough clean lanes only if you accept DFS — and at 160 MHz you simply cannot do it without two of them sharing. That is the moment to either narrow the channels or move the demanding devices to 6 GHz. The width you choose is really a decision about how many access points you can run without them fighting each other.
Roughly half of the 5 GHz band is sitting unused in most buildings. Here is why, and when you should take it.
DFS stands for Dynamic Frequency Selection. A large block of 5 GHz channels is shared with weather radar, military radar and some airport systems. Regulators allow WiFi to use those channels on one condition: if the access point detects a radar pulse, it must stop transmitting on that channel immediately and move elsewhere. That is the whole bargain.
In return you get a large number of extra channels that most consumer equipment never touches, because manufacturers avoid them for a quiet life. In a dense building those channels are often the only genuinely clean air available.
What a radar event actually looks like: the access point vacates the channel, then must listen on the new one before transmitting — a period called the channel availability check, which can take up to sixty seconds, or ten minutes on some weather-radar channels. Devices drop and reconnect. Once a quarter in a quiet area is a fair trade for clean air. Twice a day beside an airport is not, and no amount of configuration will change it.
A newer generation does not make the signal travel further. It makes better use of the signal that arrives.
| Aspect | WiFi 4 | WiFi 5 | WiFi 6 / 6E | WiFi 7 |
|---|---|---|---|---|
| Standard | 802.11n | 802.11ac | 802.11ax | 802.11be |
| Bands | 2.4 + 5 | 5 only | 2.4 + 5 (+6 on 6E) | 2.4 + 5 + 6 |
| Behaviour at weak signal | Falls off a cliff | Falls off a cliff | Degrades gracefully | Degrades gracefully |
| Many devices at once | Poor — strict queue | Poor — downlink only | Good — OFDMA both ways | Good |
| Range for the same power | Best on 2.4 GHz | Shorter | Similar, used better | Similar, used better |
| Useful at low SNR | Limited | Limited | Better — longer symbols help | Better still |
| Battery impact on devices | High | High | Lower — scheduled wake | Lower |
| Max channel width | 40 MHz | 80–160 MHz | 160 MHz | 320 MHz |
| Best used for | Legacy IoT only | Light use, few devices | The sensible default | Dense, high-demand sites |
The row that matters most is the third one. Older generations hold a fast connection until the signal weakens slightly, then collapse. WiFi 6 and 7 include modulation and coding schemes that keep working sensibly as conditions worsen. That is why an upgrade often feels like better coverage even though the physics of range has not changed at all — the edges of the coverage simply stopped falling over.
The most requested wireless fix, and the one most often applied in the wrong direction.
A survey turns opinions about the WiFi into a floor plan with numbers on it. That is the entire value.
A heatmap survey means walking a building with a floor plan loaded into software, recording signal, noise and data rates as you go, and producing a coloured map showing exactly where coverage succeeds and fails. Three kinds are worth knowing: a predictive survey models a building from drawings and wall materials before anything is installed; an active survey measures real throughput while connected; and a passive survey listens to everything in the air, including your neighbours, without joining any network.
The tool matters far less than the method. A survey walked properly with free software beats a careless one with the most expensive licence available. What you are buying is the floor plan with numbers on it — evidence that a room is at −74 dBm rather than an argument about whether the WiFi is bad. QSN surveys before quoting, precisely so the proposal describes your building instead of a generic one.
Once you can read signal, buying decisions get much simpler and usually cheaper.
Almost every wireless problem we are called to has the same shape: too few access points, turned up too high, on overlapping channels, with nothing measured. The fix is nearly always the reverse — more access points at lower power, on non-overlapping channels, placed from a survey. Three modest access points beat one powerful one in essentially every building with walls.
When comparing hardware, the specification that matters is rarely the headline speed. Look for the number of spatial streams, whether the uplink port can carry more than a gigabit if you intend to use wide channels, whether the antenna pattern suits a ceiling or a corridor, and whether the whole estate can run under one controller. Then match the generation to the building — the WiFi generations guide covers that choice, and the switches guide covers what should be feeding them.
Every one of these produces a symptom that points somewhere else.
The questions that come up once the numbers stop being mysterious.
We survey the building, map every room against the numbers in this guide, and design coverage around what we actually find — then supply, install and manage it. Across Dubai and the UAE.