Search Tools Links Login

WiFi Vision: Seeing Beyond Walls

UC Santa Barbara's Yasamin Mostofi's lab is pioneering the future of imaging. They have unveiled a revolutionary method that leverages WiFi signals to capture high-quality images of stationary objects. This groundbreaking technique utilizes the Geometrical Theory of Diffraction, particularly the Keller cones, to trace the object's edges.

Mostofi, an electrical and computer engineering professor, highlights the challenge of imaging non-moving objects with WiFi due to its stationary nature. "Instead of the traditional approach, we've zeroed in on the object's edges, offering a fresh perspective on this intricate problem," she noted. This novel approach was showcased at the 2023 IEEE National Conference on Radar.

Building on a foundation since 2009, the Mostofi Lab has consistently been at the forefront of leveraging everyday radio frequencies like WiFi for various applications, ranging from crowd analytics to smart spaces.

Mostofi elaborated on the science behind their method, "When a wave hits an edge, the Keller’s Geometrical Theory of Diffraction predicts the emergence of a cone of rays, known as a Keller cone." These cones, depending on their orientation, leave distinct patterns on a receiver grid, which are then interpreted using a specialized mathematical framework to create an edge map.

The core of their technique is the Keller cone-based imaging projection kernel. This determines the existence and orientation of edges, refining imaging accuracy. Anurag Pallaprolu, the project's lead Ph.D. student, emphasized the importance of edges, explaining a unique propagation method to enhance the image further.

Contrary to traditional methods that often produce low-quality images with WiFi transceivers, this new approach promises clarity, especially as surfaces can appear different at varying frequencies.

The lab's experiments involved using three standard WiFi transmitters and mobile receivers mimicking a WiFi grid. They successfully tested their technique in diverse environments, even managing to read the English alphabet through walls—a monumental achievement given the intricacy of the letters. In one striking demonstration, they accurately imaged and deciphered the word "BELIEVE" behind a wall, along with various other objects.

This breakthrough can potentially redefine the horizons of RF imaging, adding another dimension to what we perceive through WiFi.

About this post

Posted: 2023-09-12
By: dwirch
Viewed: 68 times




No attachments for this post

Loading Comments ...


No comments have been added for this post.

You must be logged in to make a comment.