Integrating UWB for Indoor Navigation in Mobile Apps

Precise Indoor Positioning: The Role of UWB and Its Capabilities

We integrate UWB (Ultra-Wideband) for precise positioning in mobile applications. GPS does not work indoors — that is well known. Bluetooth Low Energy offers 1-3 meter accuracy at best. Wi-Fi RSSI triangulation gives 2-5 meters with high instability. UWB — a technology with 10-30 centimeter accuracy based on measuring radio signal propagation time (Time of Flight / Two-Way Ranging). This is what Apple built into iPhone 11+ via the U1 chip, and what gives AirTag the precision of "your bag is here, turn right." Our team has 10+ years of experience in mobile solution development, and we guarantee a turnkey UWB integration. Get a consultation on selecting UWB equipment for your facility.

Why UWB Is Better Than BLE and Wi-Fi for Indoor Navigation

UWB provides accuracy 10 times better than BLE, as confirmed by our projects for shopping malls and warehouses.

How UWB Works and What It Means for Developers

UWB uses pulses ~500 MHz wide in the 6-8.5 GHz range. The signal propagation time between two devices is measured with nanosecond accuracy (TWR — Two-Way Ranging, or TDoA — Time Difference of Arrival). Distance is calculated from time: 1 ns ≈ 30 cm.

For an indoor positioning scenario, you need anchors — UWB beacons with known coordinates — and a mobile device (tag). Based on measured distances to at least three anchors, the position is calculated via trilateration.

Supported Devices

iOS: Apple NearbyInteraction framework. Devices with U1/U2 chip: iPhone 11–15, iPhone SE 3rd gen, AirTag, HomePod mini 2, Apple Watch Ultra. NISession is the main class. One session = one device pair. For positioning relative to multiple anchors, use multiple parallel NISession instances.

Android: UwbManager from Jetpack Core UWB (androidx.core:core-uwb). Supported devices: Samsung Galaxy (S21 Ultra+, S22+, S23, S24, Z Fold3+), Pixel 6 Pro+, some Xiaomi. Check support: UwbManager.isAvailable().

UWB anchors (hardware): For infrastructure positioning, third-party anchors are needed: Qorvo DWM3000EVB, Decawave DWM1001, Sewio RTLS, Pozyx. They communicate via IEEE 802.15.4z and have SDKs for configuration.

Platform Comparison: iOS vs Android

Platform Limitations

Apple Nearby Interaction is peer-to-peer only between two Apple devices or with MFi-certified accessories. For infrastructure indoor positioning (anchors → phone) directly via NISession, it works only with Qorvo-compatible anchors via a special NIConfiguration.

NISession requires exchanging NIDiscoveryToken between devices beforehand — typically via Multipeer Connectivity, Bluetooth, or a server. After token exchange, NISession.run(configuration:) starts measurements.

How to Integrate UWB into a Mobile App?

Practical Case: Navigation in a Shopping Mall (from Our Practice)

Scenario: a shopper looks for a specific store. GPS is unavailable. BLE navigation is not accurate enough for 3-meter-wide corridors. UWB anchors are installed on the ceiling every 10-15 meters.

Pozyx Creator anchors (UWB, PoE, self-localization) → central Pozyx server collects positioning data → REST API provides tag device coordinates in the building's coordinate system.

Mobile app: upon entering the building, the device "connects" to the system (via BLE handshake for identification), then every 100-200 ms receives coordinate updates via WebSocket (x, y, floor).

Coordinates are overlaid on the building plan (SVG floor scheme). Smooth marker movement: Kalman filter to smooth noisy UWB measurements. Without the filter, the marker "jumps." A Kalman filter on the mobile device is 20-30 lines of code but significantly improves UX.

Navigation to a point: A* pathfinding on a graph of passages (the graph is built from the SVG scheme, with forbidden zones — walls and display cases). If deviation from the route exceeds 1m, recalculate.

Integration with Apple NearbyInteraction

For device-to-device scenarios (courier → client, warehouse worker → specific pallet):

import NearbyInteraction class UWBSession: NSObject, NISessionDelegate { let session = NISession() func startSession(with peerToken: NIDiscoveryToken) { session.delegate = self let config = NINearbyPeerConfiguration(peerToken: peerToken) config.isCameraAssistanceEnabled = true // iOS 16+: AR overlay session.run(config) } func session(_ session: NISession, didUpdate nearbyObjects: [NINearbyObject]) { guard let peer = nearbyObjects.first else { return } if let distance = peer.distance { print("Distance: \(distance) m") } if let direction = peer.direction { // SIMD3<Float> - direction in 3D print("Direction: \(direction)") } } } 

isCameraAssistanceEnabled enables Precision Finding — an AR arrow over the camera shows direction to the object (like in AirTag Precision Finding). Requires ARKit and NSCameraUsageDescription.

Token Exchange

NIDiscoveryToken cannot be created programmatically — only obtained from session.discoveryToken. To start a UWB session, both devices must exchange tokens beforehand. Typical scheme: both devices publish the token via Bluetooth Peripheral → scan each other → receive tokens → start NISession.

Using CloudKit or a server — for scenarios where devices are not physically close during initialization.

Typical Integration Problems

• Multipath interference. UWB signal reflects from metal surfaces (shelves, equipment) — false distance measurements. Solution: NLOS (Non-Line-of-Sight) detection via analysis of First Path Power vs Total Received Power. Pozyx and Decawave return these metrics in the raw packet.
• NISession suspended. iOS pauses the UWB session when the app goes to background. sessionWasSuspended(_ session:) — save last position; on sessionSuspensionEnded, restart. UWB does not work in background — platform limitation.
• Anchor calibration. Anchor coordinates in space must be measured precisely — a 5 cm error shifts all calculations. Self-localization of anchors (Pozyx, Sewio) automatically determines their coordinates at first startup via UWB TWR between themselves.
• Accuracy in motion. At high speed (> 2 m/s), TDoA systems give more errors than TWR. For pedestrians, TWR with 10 Hz update is sufficient.

What Is Included in the Work?

• Audit of infrastructure and use scenarios
• Selection of UWB platform and equipment
• SDK integration (iOS NearbyInteraction, Android UwbManager, anchors)
• Kalman filter and navigation graph development
• Load testing (up to 100+ devices)
• Documentation, support, and team training

Project Process

1. Audit of use scenario and equipment
2. Selection of UWB platform (Apple NI / Qorvo / Pozyx)
3. Pilot on a test zone with accuracy measurement
4. Integration with the mobile app
5. Kalman filtering and navigation graph
6. Load testing (100+ simultaneous devices)
7. Anchor deployment and commissioning

Timeline

A pilot with Apple NearbyInteraction on two devices takes 1-2 weeks. A full indoor navigation system with infrastructure anchors, Kalman filter, and building map takes 2-4 months depending on the facility's scale. The cost is calculated after assessing the infrastructure and target scenarios.

Contact us for a consultation on your project. Order a turnkey UWB integration — get precise indoor positioning.