Coded environments: data-driven indoor localisation with reconfigurable intelligent surfaces

Abstract

Reconfigurable Intelligent Surfaces have recently emerged as a revolutionary next-generation wireless networks paradigm that harnesses engineered electromagnetic environments to reshape radio wave propagation. Pioneering research presented in this article establishes the viability of Reconfigurable Intelligent Surfaces-enhanced indoor localisation and charts a roadmap for its integration into next-generation wireless network architectures. Here, we present a comprehensive experimental analysis of a Reconfigurable Intelligent Surfaces-enabled indoor localisation scheme that evaluates the localisation accuracy of different machine learning algorithms under varying Reconfigurable Intelligent Surfaces states, antenna types, and communication setups. The results indicate that incorporating Reconfigurable Intelligent Surfaces can significantly enhance indoor localisation accuracy, achieving an impressive 82.4% success rate. Moreover, this study delves into system performance across varied communication modes and subcarrier configurations. This research is poised to lay the groundwork for implementing Reconfigurable Intelligent Surfaces-empowered joint sensing and communications in future next-generation wireless networks. Syed Tariq Shah and colleagues use multi-antenna reconfigurable surfaces to maximise the accuracy of wireless indoor localisation. They study the achievable performance improvement using pre-trained machine learning techniques.