Poster: Using Commodity WiFi Devices For Object Sensing And Imaging

Abstract

Object identification and imaging play an important role in many real-life applications such as robotics, automated vehicle networks and search and rescue operations in the aftermath of natural disasters. Existing traditional imaging systems require installing custom-built hardware and dedicated infrastructure which are expensive and not scalable. Also, they are not occlusion immune. With the pervasive and wide availability of WiFi infrastructure, in this project, we explore the possibility of seeing the world through the low-priced commodity WiFi devices by exploiting multipath reflections. WiFi-based solutions are promising due to their ubiquity and low cost. And unlike optical and infrared signals, WiFi can “see-through” walls, clothes and fabrics. We prototyped a $ 6 \times 6 -$antenna planar array using commodity Intel NUCs and used 4 different objects for creating an image using a 2D Fourier transform. Our initial results using commercial off-the-shelf (COTS) hardware show promising results in the Line of Sight (LOS) environment.