Human activity monitoring is crucial for assisting independent living among the elderly and disabled. While microwave sensing offers a privacy-preserving alternative to cameras and wearables, existing techniques struggle in complex, non-line-of-sight (Non-LOS) environments. This paper introduces Intelligent Wireless Walls (IWW), using reconfigurable intelligent surfaces (RIS) and machine learning to enable high-precision activity monitoring in challenging settings. Experiments in corridor junction and multi-floor scenarios demonstrate a significant accuracy gain (up to 28%) compared to traditional microwave sensing, showcasing the potential of IWW for enhancing contactless in-home monitoring.

Muhammad Usman, James Rains, Tie Jun Cui, Muhammad Zakir Khan, Jalil ur Rehman Kazim, Muhammad Ali Imran, Qammer H. Abbasi