WiFi Signal Strength Explained

"Two bars" means something different on every phone, laptop, and router — WiFi signal reporting has no universal standard, which makes the underlying dBm measurement far more meaningful once you know how to read it. This post covers what RSSI and SNR actually measure, why bigger negative numbers mean weaker signal, and why estimating distance from signal strength alone only ever produces a rough approximation.

RSSI: a logarithmic, negative scale

RSSI (Received Signal Strength Indicator) is measured in dBm — decibels relative to one milliwatt — on a logarithmic scale, which is why the numbers look unintuitive at first: −30 dBm is a very strong signal, −90 dBm is very weak, and the scale runs negative because WiFi signal power is almost always a small fraction of a milliwatt by the time it reaches a receiver. Each 3dB step roughly doubles or halves the actual signal power, which means the difference between −60 and −70 is far larger in real terms than the flat 10-unit gap suggests.

Why there's no universal RSSI-to-percentage standard

Every operating system, driver, and app that shows a signal percentage or a bar count applies its own conversion formula — there's no IEEE or industry standard defining exactly how −65dBm should map to a specific percentage. This is why the same physical connection can show "3 bars" on one device and "70%" on another simultaneously; comparing raw dBm values (available in most WiFi analyzer apps) is the only genuinely apples-to-apples comparison across different devices.

SNR: strength relative to noise, not strength alone

Signal-to-Noise Ratio compares your actual signal strength against the background RF noise on the same channel — other WiFi networks, Bluetooth devices, microwave ovens, baby monitors, and general electromagnetic interference all raise the effective noise floor a receiver has to distinguish your signal from. A −60dBm signal against a −90dBm noise floor (30dB of SNR) performs meaningfully better than the identical −60dBm signal against a −65dBm noise floor (only 5dB of SNR) — the RSSI alone doesn't reveal this difference at all.

SNR        Rating       Typical experience
40dB+      Excellent    Maximum throughput, very reliable
25-40dB    Good         Solid performance for most uses
15-25dB    Fair         Usable, some slowdowns possible
10-15dB    Poor         Noticeable issues, frequent retries
Under 10dB Unusable     Connection barely functions, if at all

Estimating distance: the log-distance path loss model

Signal strength predictably weakens with distance, following (approximately) the log-distance path loss model: signal strength drops by a fixed amount each time distance from the transmitter doubles, with the exact rate controlled by a path loss exponent — around 2 in open space with no obstructions, 3 to 4 indoors where walls and furniture absorb and scatter the signal. Rearranged to solve for distance given a reference signal strength at a known distance (typically 1 meter), this gives a genuinely useful but inherently approximate distance estimate.

The word "approximate" matters here: multipath reflection (the same signal arriving via several different bounced paths and interfering with itself, constructively or destructively), specific wall materials, furniture, and even people moving through the space all affect real-world signal strength in ways the simplified model doesn't capture. It's useful for a rough sense of distance or comparing relative readings, not for precise positioning.

The sticky client problem

A device roaming through a home or office with multiple access points ideally switches to whichever AP currently gives it the strongest signal — but many devices are conservative about roaming, staying connected to an already-weak AP well past the point a nearby one would serve them better, simply because the connection hasn't fully failed yet. This is commonly called the sticky client problem. Standards like 802.11k (which lets an AP tell a client about neighboring APs) and 802.11v (which lets an AP suggest a client roam) exist specifically to make this handoff smarter, though support and actual behavior vary considerably across device manufacturers — measuring RSSI from a specific device's own perspective is often the most direct way to confirm whether it's actually experiencing this problem at a given location.

Reading RSSI directly from your own device

# Windows
netsh wlan show interfaces

# macOS
system_profiler SPAirPortDataType

# Linux
iw dev wlan0 link

Each of these reports the current connection's live RSSI (and often noise level too) directly from the WiFi driver — the actual raw number behind whatever bars or percentage your OS displays elsewhere, and the same value this calculator's quality and SNR figures are computed from.

A single reading is a snapshot, not a constant

RSSI fluctuates continuously even from a stationary device in a stationary location — multipath effects shift as anything in the environment moves, and interference from other devices comes and goes. A meaningful assessment of signal quality in a specific spot benefits from several readings over time rather than trusting a single measurement as definitive.

Common mistakes

• Comparing signal bars or percentages across different devices. Neither is standardized; compare raw dBm values instead.
• Treating strong RSSI as a guarantee of good performance. A strong signal on top of an equally strong noise floor (low SNR) still performs poorly.
• Trusting a distance estimate as precise. The log-distance model is a genuinely useful approximation, not a measurement.
• Drawing conclusions from a single signal reading. Real RF conditions fluctuate continuously; one measurement is a snapshot.

FAQ

Try it yourself

WiFi Signal Strength Calculator converts an RSSI reading into a quality percentage, SNR rating, and estimated distance, entirely in your browser.