Next2Me: Capturing Social Interactions through Smartphone Devices using WiFi and Audio signals

Typical approaches in detecting social interactions consider the use of co-location as a proxy for real-world interactions. Such approaches can underperform in challenging situations where multiple social interactions can occur in close proximity to each other. In this paper, we present a novel approach to detect co-located social interactions using smartphones. Next2Me relies on the use of WiFi signals and audio signals to accurately distinguish social groups interacting within a few meters from each other. Through a range of real-world experiments, we demonstrate a technique that utilises WiFi fingerprinting, along with sound fingerprinting to identify social groups. Experimental results show that Next2Me can achieve a precision of 88% within noisy environments, including smartphones that are placed in users’ pockets, whilst maintaining a very low energy footprint (<3% of battery capacity per day).