We recorded the ROI already in the first month with a 40% increase in product revenue!
How it works
How Hybrid Fusion works
Hybrid Fusion is Ariadne's patented signal-based people counting method, built for physical venues without cameras or facial recognition. It combines phone signal detection with Time-of-Flight depth sensing on a single edge device, then fuses both streams into counts, dwell times, and movement paths. No identifier is captured along the way. Identifiers are stored only when a visitor explicitly opts in.
Signal sensing
Patented phone signal sensing
Phones constantly broadcast signals as they search for connections. Ariadne's patented sensing detects every signal a device emits, even in airplane mode, and triangulates location to within 30 cm. No image. No video. No MAC address captured by default. Identifiers are stored only when a visitor explicitly opts in (for example, through a guest Wi-Fi login).
Works even when phones are in airplane mode
- Detects every emitted signal across the venue
- 30 cm distance accuracy per detection
- No MAC capture by default; opt-in only
- Sees journeys, not faces
Depth sensing
Time-of-Flight depth sensing
At every entry, exit, and choke point, Time-of-Flight units fire infrared light pulses and measure the return distance to whatever crosses the beam. Counts are device-independent: phone, no phone, child in a stroller, every body counted.
- Counts every visitor regardless of device
- Sub-second update rate, holds in dense crowds and low light
- Captures geometry, not images. No video, no still frames, no biometric data
- 30 cm accuracy at venue entrances
Why it's called Fusion
Two streams, one fusion layer
Both sensing technologies share the same hardware. Signal sensing handles the venue interior. Time-of-Flight handles entries. The sensor streams both feeds to Ariadne, where Hybrid Fusion combines them into a single trajectory per visit and computes counts, dwell, and paths. The streams leaving the sensor carry no identifier: no MAC, no device ID, no biometric data.
Centralised processing
Sensors capture and stream events. Fusion, computation, and analytics all happen centrally in Ariadne. No per-venue server to provision.
Identifier-free by design
Outputs are counts and paths. No identifier is captured in the first place, so there is nothing to anonymise after the fact.
Single integration
One sensor, one event stream into Ariadne, one dashboard. No camera, Wi-Fi, or BLE network to stitch together.
From emission to insight
How a visit becomes a data point
1
Step 01
Emit
Phones broadcast Wi-Fi probe requests, Bluetooth and BLE advertisements, and cellular housekeeping signals as they search for networks. Ariadne's patented signal sensing captures all of these radio types across the venue, even on devices in airplane mode.
2
Step 02
Detect
Multiple Ariadne sensors across the venue receive every probe within range and timestamp each arrival. Multilateration on time-of-arrival yields 30 cm position accuracy per detection. No identifier is stored by default.
3
Step 03
Range
Time-of-Flight units at entries and choke points fire infrared light pulses and measure the round-trip time of each reflection to compute depth. Every body crossing the beam is counted by its silhouette: adults, children, strollers, independent of any phone. The sensor captures geometry, not appearance: no frames, no faces, no biometric data.
4
Step 04
Stream
The sensor sends timestamped signal and depth events to Ariadne. The events leaving the sensor carry no identifier because none was captured.
5
Step 05
Fuse
Hybrid Fusion combines the two event streams into a single trajectory per visit, then computes counts, dwell, and paths in the SaaS.
6
Step 06
Present
Dashboards show counts, dwell time, paths, and conversion in real time. Only metrics flow back to the venue operator: no personal data, because none was ever captured.
Explore the dashboard30cm
Detection precision
Sub-meter accuracy across the venue
Hybrid Fusion delivers 30 cm accuracy on every detection, then fuses depth and signal into the same anonymous trajectory. No cameras, no facial recognition, no MAC capture by default.
See it for your venuePrivacy by architecture
GDPR-native, EU AI Act ready
Ariadne does not collect MAC addresses or device IDs by default. Identifiers are captured only when a visitor explicitly opts in (for example, through a guest Wi-Fi login). The architecture sits structurally outside Annex III high-risk categories of the EU AI Act because no biometric data is captured or processed. GDPR-native by design (Article 25).
No cameras
No video, no still frames, no facial-recognition pipeline.
No biometrics
Ariadne does not measure faces, gait, age, or gender. Anti-feature by design.
No PII by default
Counts and paths are anonymous. Identifiers exist only on visitor opt-in.
Audited and certified by
Outputs
What lands in the dashboard
Hybrid Fusion outputs four families of metric. Each is venue-wide, anonymous, and updated in real time.
Counts
Entries, exits, occupancy by zone.
Dwell
Time spent per zone, queue length, wait time.
Paths
Identifier-free journey from entrance to exit, with hand-off points between zones.
Conversion
Entries vs intended action: purchase, gate, departure.
Customers
Proof, from real venues
Specific outcomes, not slogans. Each story links to the full deployment.
When they said: We are the Google Analytics of the physical world, they were not kidding!
A truly plug-n-play solution! I did the installation myself!.
Two ways in
Plug in our kit, or use yours
Hybrid Fusion runs on Ariadne sensors at venues that want a clean install, or on EaseLink at venues that want to use existing Wi-Fi access points and people-counting sensors (Aruba, Cisco, Sophos, Ruckus, TP-Link APs plus Milesight and Xovis 3D/ToF sensors). Both paths feed the same identifier-free analytics.
Works with
Originated in research
Built on patents, not slogans
Hybrid Fusion grew out of work at the Technical University of Munich and is protected by patents granted across the EU, Germany, and Canada. The German patent specifically covers combining radio signal strength with electromagnetic-radiation monitoring to refine indoor positioning. That is the core method behind signal-based people counting at Ariadne.
DE102022115597A1
Method for Improving Accuracy of Indoor Positioning and System for Position Estimation of an Individual or Object
Germany · 2022
The Hybrid Fusion technique itself: combining radio signal strength with electromagnetic-radiation monitoring to refine location.
EP3964852A1
Method and system for providing a user carrying a mobile device with navigation and navigation services
European Patent Office · 2020
The browser-native, no-app indoor navigation that powers EaseLink.
CA3125501A1
Method and system for tracking a mobile device
Canada (TU Munich) · 2020
The particle-filter tracking algorithm that anchors Ariadne's positioning math.
Inventors: Georgios Pipelidis, Nikolaos Tsiamitros (Ariadne co-founders), with Nam Le Duc, Tobia Mancabelli, and Christian Prehofer (Technical University of Munich).
FAQ
How Hybrid Fusion works, in detail
Quick answers on the technology layer.
What is Hybrid Fusion?
Hybrid Fusion is Ariadne's patented method for measuring people in physical venues by combining phone signal detection with Time-of-Flight depth sensing. The sensor streams identifier-free events to Ariadne, where the two streams are fused into trajectories and computed into counts, dwell times, and movement paths. No identifier is captured in the first place, so nothing personal flows through the pipeline. No cameras, no facial recognition, no MAC capture by default.
How does signal sensing work in airplane mode?
Phones broadcast low-level probe signals to maintain radio housekeeping even when the user has enabled airplane mode. Ariadne's signal sensor receives those probes and uses time-of-arrival to triangulate position to within 30 cm. The sensor never reads or stores the MAC address by default.
Is the data really anonymous?
Yes, structurally. By default Ariadne does not capture MAC addresses or device IDs at all. The signal sensor reads radio probes for time-of-arrival; identifiers are never read or stored unless a visitor explicitly opts in (for example, through a guest Wi-Fi login). There is nothing to anonymise after the fact because no personal data is captured in the first place.
Does it count people who don't carry phones?
Yes. Time-of-Flight units count every body crossing an entry beam regardless of whether they carry a phone. Signal sensing handles the inside-the-venue journey for the visitors who do carry one. The fused record reconciles both, so children, staff carrying multiple phones, and phone-free visitors are all represented correctly.
How accurate is sub-meter accuracy?
Ariadne hardware delivers 30 cm accuracy per detection across the venue. End-to-end accuracy depends on the venue layout, sensor placement, and how the venue defines a count (entries only vs zone-by-zone). Real-world deployments at airports and shopping centres consistently measure within 1 to 2 percent of ground truth.
Is Ariadne EU AI Act compatible?
Yes. Ariadne's architecture sits structurally outside Annex III high-risk categories because no biometric data is captured or processed. The patented signal sensing measures presence and movement without collecting MAC addresses or device IDs by default. GDPR-native per Article 25.
Where to next
Ready to see it for your venue?
Talk to us
Two questions, twenty minutes, a real walkthrough of your venue's footfall.
What to expect
- 20-minute screen share, walked through on your venue map
- Live walkthrough of Hybrid Fusion sensor outputs
- Where Ariadne fits, and where it doesn't
Got a different question?
Send us a message
Anything that isn't a sales conversation. We'll route it to the right person and get back within one business day.