Intelligent Metasurface Based Indoor Positioning and Control System With Non-Cooperative Blind Signals

摘要

It is ever-increasingly demanded to monitor the position of the people and the robot in industrial production and home life scene. However, conventional systems are limited in use because of low location accuracy and huge energy consumption. To utilize the ambient potential resources like the Wi-Fi signal, we propose a novel indoor positioning system simultaneously, uses the artificial neural networks (ANNs) to extract the necessary information for positioning and imaging. We use the metasurface to allocate the distribution and control the propagation of the Non-Cooperative Blind Wi-Fi signal to achieve the communication with the intelligent terminal. Other ANNs like deep Q-learning structure network (DQN) for comprehending the information and controlling the indoor intelligent entities are employed. Instantaneous in-situ positioning and imaging of full scene and real-time instruction issue of multiple intelligent entities have been experimentally demonstrated. The reported strategy could open a new avenue for future smart factory, smart homes, internet of things, healthy monitoring, and safety screening free of visual privacy issues.