Research built around difficult questions.

Industrial warehouses are difficult environments for conventional positioning systems. Fixed reference nodes create costs across hardware, wiring, power, networking, calibration, installation and maintenance.

This research extends the Protraction Technique into three-dimensional industrial environments, investigating whether devices can reconstruct spatial relationships using multi-level binary wireless connectivity rather than conventional ranging.

The work moves localization toward systems that require significantly less positioning infrastructure.

Wireless signals contain information about location, but extracting that information accurately can be difficult.

This research investigated machine-learning techniques for wireless location prediction, exploring how learned relationships between wireless observations and physical positions could improve localization.

It represents an important bridge between traditional wireless positioning and increasingly intelligent localization systems.

Much of the later research direction began with a deceptively simple industrial problem: How can physical assets become visible to software?

This research explored wireless sensor networks capable of combining location information with environmental sensing for industrial automation.

It helped establish the direction that would later expand into wireless positioning, machine-learning localization and infrastructure-light spatial intelligence.