Behavior Sensing Infrastructure Project

Behavior Sensing Infrastructure Project

Research Lead: Hajime Nagahara, Professor, Institute for Datability Science, Osaka University

The Behavior Sensing Infrastructure Project endeavors to build real-world behavior sensing infrastructure by collecting a variety of signals such as Wi-Fi signal reflection intensities, spectrograms of channel state information (CSI), GPS, acceleration/angular velocity information and image/audio information from indoor and outdoor sensors, including Wi-Fi routers, sensors attached to wearable devices such as smartphones and smartwatches, security cameras, and laser rangefinders.

Using the above signals as input, we develop modules for low-order functions necessary for behavior sensing. Eventually, we plan to build a variety of behavior sensing modules. One example is a module for detecting/tracking humans based on analysis of security camera images. Another example is a location estimation module using a combination of GPS, inertial sensors (e.g., accelerometer), and/or from Wi-Fi signal strength information.