Coverage, measured. Not assumed.

Your network says covered. We fly over and check.

SignalWings measures what your IoT network actually delivers, report by report, and surveys the coverage nobody has mapped: on foot, by vehicle, and from the air with sensor-carrying pigeons.

Book a 2-week pilot Free network check

Grand Finalist EIT Jumpstarter 2025
Featured Kurrant · Nature as a Sensor

Illustration · each bar is one hour of expected reports

The silent failure

Online is not the same as delivering.

Status dashboards answer one question: can we reach the device? They never ask whether the readings it promised actually arrived. A sensor can show green all week and still lose a third of its data.

OnlineWhat the status page shows

71% deliveredWhat actually arrived

Silent gaps

Reports stop for hours while the link stays up. We measure the longest gap against the cadence each device promised.

Frozen values

The device keeps sending the same number. Delivered, technically. Useless in practice, and invisible on a status page.

Out-of-range readings

Values outside what the declared metric can physically be. That is a broken probe, not a heatwave.

Coverage that was never there

Zones on the floor plan that no gateway actually hears. A device list can't show what was never installed.

The console

One look at which devices are lying.

Every device carries an expected reporting interval and a declared metric. SignalWings scores what share of expected reports arrived, separately from link health, so “degraded” and “delivering” stop being confused.

Console overview: share of expected reports arriving, blind spots, coverage risk and delivery shortfall by hour — One sentence first: how much of what was expected actually arrived, and what needs a decision today.Demo console · synthetic data

Delivery vs. cadence: Share of expected reports that arrived, per device, zone and site.
Link health kept separate: Status, RSSI and uptime never stand in for delivered data.
Measured or inferred: Every finding says whether it was measured on site or estimated from nearby samples.
Coverage score per zone: Each zone carries its own score, dead zones and the survey that produced them.

Coverage surveys

Coverage you can't infer, you survey.

A device list tells you what you installed. A survey tells you what actually works across the area. Every assessment ends with a heatmap, dead zones, a confidence level and findings marked measured or inferred.

Walk survey

Handheld receivers carried through the site. Best for warehouses, plants, basements and anything indoors.

Indoor · mixed

Vehicle survey

Drive-through measurement for corridors, campuses, industrial parks and distributed meters.

Outdoor · large areas

Pigeon wearing a lightweight sensor harness, held by a handler

Flight survey

Sensor-carrying pigeons cover rooftops, courtyards and the space between buildings: the 3D layer street-level surveys miss.

Outdoor · urban · welfare-gated

01Requested
02Triaged
03Scheduled
04In field
05Processing
06Report ready
07Delivered

Bio-IoT flight survey

The fastest way to cover a city block is to fly it.

Racing pigeons already cross the rooftops, yards and gaps between buildings that walk and vehicle surveys can't reach. We fit them with a lightweight LoRa sensor and measure real signal in three dimensions.

“If you want to cover an area quickly, you can simply fly — and what better way than with the animals that are already there?”
Stefan Bjelaković, CTO · interview on Kurrant

How flight surveys work

Two team members fitting a sensor harness on a pigeon

The sensor module with LoRa radio and antennas

Welfare gate No check, no flight. Scheduling stays blocked until all five pass.

Payload weightSensor and harness checked as a share of each bird's body weight before it flies.
Veterinary sign-offA vet clears every carrier for every run.
Weather clearanceWind, rain and visibility inside safe limits for the whole flight window.
Temperature limitsNo flights in heat or cold outside the agreed range.
Permit on recordEach run references its permit before it can be scheduled.

Where it pays off

Built for sites where a missing reading costs money.

Findings are written the way an operator would say them, and each one names the measurement behind it.

Warehouses & logistics

Forklifts, mobile equipment and operational zones. Find where assets disappear and which aisles no gateway hears.

Forklift not seen in expected zone
Gateway coverage weak near loading dock

Industrial sites

Distributed sensors and gateways across plants and industrial parks, where a missed reading becomes a missed fault.

Sensor cluster frozen since shift change
New blind spot after racking moved

Utilities & smart metering

Validate radio coverage and gateway placement for meters and field sensors before and after rollout.

Planned vs. measured coverage mismatch
Gateway visibility inconsistent over 24 h

Smart cities

Mobility, environment and public systems in dense streets, where buildings bend and block the signal.

Dead zone behind dense block
Instability along transport corridor

Pilot · 2 weeks

Start with two weeks and one honest map.

The pilot is built for a fast a-ha moment: a clear picture of what is active, missing, weak or quietly failing on one site, before you commit to anything bigger.

Book a pilot

Day 1
Discovery workshop
We walk the site with your team, review what's installed and pick the zones that matter most.
Days 2–5
Deploy and survey
Sensors and gateways go in or get integrated. Walk, vehicle or flight survey runs across the area.
Days 6–13
Monitor delivery
Every device is scored against its expected cadence: gaps, frozen values, out-of-range readings.
Day 14
Findings and next steps
Coverage heatmap, dead zones, anomaly report and a ranked list of what to fix first.

Free · 2 minutes · no call

Is your sensor network lying to you?

Eight questions about how you monitor today. You get a score across four areas, the blind spots it points to, and the Silent Failure Field Guide by email.

Start the network check Just send me the guide →