Decoupled UWB Sensing and Sub GHz Telemetry Architecture for Indoor Localization

Project description

The main research challenge is evaluating a decoupled indoor localization system that separates location sensing from data transmission in multi story concrete buildings. It asks whether high precision Ultra Wideband sensors can be reliably combined with an 868 MHz Sub GHz telemetry backbone to track locations across multiple floors.

The design must overcome severe signal blocking by walls and human bodies, while strictly adhering to European 1 percent duty cycle transmission limits and preventing network congestion.

Context

This project operates in the healthcare domain, focusing on resident tracking in nursing homes. Institutions need reliable tracking to support caregiving staff, but multi story concrete buildings create major technical issues. Current 868 MHz tracking systems easily penetrate walls to save infrastructure costs, but this causes signal leakage between floors, leading to location errors of up to 17 meters.

On the other hand, highly accurate sensors like Ultra Wideband are easily blocked by walls and require very expensive, dense wiring to cover a whole building.

This project explores a middle ground. It combines the precise, room limited sensing of UWB with the wall penetrating data transmission of 868 MHz to create an affordable and accurate medical tracking system.

Results

The primary outcome is the validation of the decoupled architecture, which successfully solves the issue of tracking patients on the wrong floor. Physical testing proved that UWB sensors meet the 1 meter medical accuracy requirement inside a room, but their signals are completely blocked by concrete ceilings. This structural blocking is actually a benefit, as it guarantees a patient is physically on the reported floor.

A major insight is that fixing the spatial accuracy shifts the system bottleneck to the network layer. Mathematical models showed that European transmission laws limit the anchors to sending one location update every 4.6 seconds. Furthermore, simulations revealed that when multiple nodes transmit randomly, airwave collisions cause data delivery to drop below the 90 percent medical safety threshold at around 14 active anchors.

The project sits at Technology Readiness Level 4. The basic technological components have been integrated and validated through hardware testing and simulation in a realistic indoor environment. It provides clear value by showing the stakeholder that the physical hardware architecture is highly viable, but that moving to higher TRL stages will require custom network scheduling software to prevent data congestion.

About the project group

Petar is a HBO-ICT graduate, with a specialization in Infrastructure. The project took one full-time semester to execute and was done individually, but in a group setting, where feedback could be exchanged.