Real-Time IoT Fusion
Sensors publish, the twin updates. High-frequency telemetry lands in a time-series store and surfaces at the exact position in 3D space where the reading was taken — not in a separate dashboard you have to mentally reconcile.
How it works
Connect
Sensors publish to an MQTT broker over TLS. Gateways, PLCs, and existing SCADA systems bridge in the same way, so hardware you already own does not need replacing.
Store and index
Readings are written to InfluxDB, a time-series database built for high-frequency writes. Asset and organisation metadata lives alongside it in Firestore.
Surface in the twin
The API rebroadcasts over WebSocket to the front end, so a temperature spike appears on the right pump in the 3D scene rather than as a row in a table.
Capabilities
Time-series foundation
InfluxDB is purpose-built for rapid sensor writes — vibration analysis, thermal readings, flow rates, and environmental metrics — without the write amplification a relational store would hit.
MQTT and WebSockets
Sensors stream over lightweight MQTT, ideal for constrained devices and unreliable links. The platform rebroadcasts to browsers over WebSocket for genuinely live dashboards.
Spatial telemetry
Every sensor is bound to a position inside the twin. Readings render where the hardware physically sits, which turns an anomaly into a location instead of a device ID.
Thresholds and alerts
Define bounds per sensor and raise alerts when readings breach them. Alerts carry their spatial context, so responders know which asset and which floor before they arrive.
Authenticated end to end
Every sensor, drone, and user authenticates through Firebase Auth with role-based access control, over TLS. Data is scoped by organisation at the API layer.
Open integration
REST APIs for platform data and MQTT/WebSocket for real-time streams mean telemetry can also flow outward into the BI and maintenance systems you already run.
Under the hood
- Ingest
- MQTT over TLS; gateway, PLC, and SCADA bridging
- Time series
- InfluxDB for high-frequency sensor readings
- Metadata
- Firestore for asset, sensor, and organisation records
- Realtime out
- WebSocket push to browser and mobile clients
- Auth
- Firebase Auth with role-based access control, TLS throughout
- APIs
- Versioned REST endpoints under /v1
Frequently asked questions
What does IoT fusion into a digital twin actually mean?
It means each sensor reading is bound to a position in the 3D model rather than living in a separate dashboard. A vibration anomaly shows up on the specific pump that produced it, so diagnosis starts from location and asset context instead of from a device identifier in a spreadsheet.
Which protocols and sensors are supported?
Sensors publish over MQTT, the standard lightweight protocol for constrained devices and unreliable networks. Existing gateways, PLCs, and SCADA systems bridge into the same broker, so most industrial hardware already in place can be connected without replacement.
Why InfluxDB rather than a relational database?
High-frequency telemetry is append-heavy and time-ordered, which is exactly what a time-series database is optimised for. InfluxDB handles rapid sensor writes and time-window queries efficiently, while asset and organisation metadata stays in Firestore where relational-style lookups make more sense.
How is sensor data secured?
Sensor, drone, and user traffic is authenticated through Firebase Auth with role-based access control and carried over TLS. API endpoints enforce organisation scoping, so telemetry from one tenant is never returned to another.
Can telemetry feed predictive maintenance?
Yes. Because readings are retained as time series rather than only as current values, degradation trends are visible over time. Miyamoto AI queries those streams to surface wear patterns and thermal anomalies ahead of failure.
Explore the rest of the platform
See it on your own site
Start a 30-day trial, or have KatanaTech operators run the capture and hand you a finished twin.