The research poster where it all began. This early prototype introduced a wireless location system that needs no fixed cabling and no synchronised clocks, the first step of a decade of ranging research.
The motivation
Radio-frequency identification and tracking are everywhere: managing assets, securing sites, and following the movement of people through specialised environments. But the high-accuracy positioning techniques of the day relied on time-difference-of-arrival ranging with fixed reference nodes, which in turn required wired infrastructure between references for data transfer and timing synchronisation. That cabling constrains where such systems can realistically be used.
The project set out to remove that constraint: to build an RF locationing system that works without wired infrastructure or timing synchronisation, opening the door to both fixed-reference and fully relative locationing between mobile nodes.
The prototype
The aim was a narrow-band RF time-of-flight system with sub-metre positioning resolution, suitable for locating nodes inside wireless sensor networks. It was built on a Chipcon CC2431 development kit, a fully integrated IEEE 802.15.4 transceiver paired with an Intel 8051 microcontroller.
- Software-only. The algorithm runs entirely in software, needing no additional hardware to operate.
- Relaxed synchronisation. Ranging is performed on a single 250 kb/s channel in the 2.4 GHz ISM band, without tight transceiver-to-transceiver synchronisation.
- Fixed or relative references. The architecture supports a controller with fixed reference nodes, but is designed to extend to relative locationing too.
What it demonstrated
The poster shows real-time position estimates for a "blind" node placed among reference nodes, in both line-of-sight and non-line-of-sight conditions. A brute-force graphical display renders the estimate as a heat map: warmer colours mark positions closer to the node's true location, with a single white dot marking the best estimate. It is a vivid, intuitive picture of a radio network working out where its own members are.
Where it led
This 2008 prototype was the opening chapter of a sustained body of research. Its ideas were refined into a peer-reviewed journal paper and ultimately a full doctoral thesis on radio frequency ranging: the timing, signal-processing and radar expertise that Norse Radar is built on today.
Presented through the Pervasive Systems Centre and the Electronic Systems and Devices Group, School of Electronics and Computer Science, University of Southampton.
