Image Frame Rate Limited Sub-Centimeter Localization System for Hybrid Radio-Optical Wireless Communications

Abstract

Infrared optical wireless communication (IR-OWC) has received tremendous attention for the high-speed oriented applications. In consideration of issues like power budget and privacy, narrow-width and directional optical beams are commonly applied for IR-OWC. As only users located within specific areas can get the spatially confined optical signals, accurate localization (sub-)systems are required for the central controller to determine the beam-steering strategies. Image-based localization technologies have shown great accuracy performance. However, their localization speed is typically limited by the image transportation delays and processing time. Systematic studies for the phenomena and mechanisms behind the time relevant issues are desired but still rarely reported. Many real-time localization systems and complexity-reduced computation schemes have been published. However, most of the reported systems are performed under rather controlled (ideal) environmental conditions, and issues like deployments and data transportation are commonly neglected. In this paper, a real-time localization testbed for hybrid radio and optical wireless communications is studied. Novel two-stage processes combining quasi-synchronization and local user tracking processing are proposed to increase the localization speed. Comprehensive discussions for the time relevant issues are performed. The detected images are transported through WiFi; hence the localization systems can be flexibly deployed and be easily integrated with the IoT networks. The testbed is implemented under complex environmental conditions to more accurately reveal the practical performance. An overall localization delay of 39.3 ms, which is close to the camera frame rate limitation (29.2 ms), with 0.11 cm localization accuracy is demonstrated for a 1-cm-aperture user device (UD) 180 cm away from the camera.