Researchers Use WiFi to Count the Number of Individuals in a Crowd

Go to any crowded place, like a restaurant or retail store, and try to count how many individuals are in the room. You’re either going to put a significant amount of effort into tallying the group, or you’re going to give up in frustration as people move about and you lose track of them. Still, there are many situations when it’s useful to know how many people are in a crowd. This new technique from University of California Santa Barbara researchers can accurately count them using just off-the-shelf WiFi devices, and they don’t even need to be in the same room.

There are a number of reasons why you might want to know the number of individuals in a room. Those include security, safety, and even marketing. It’s easy to imagine why a retail chain would be interested in how many people are in each of their stores at any given time, or why the TSA would like to know how many people are at an airport gate. The technique developed at UCSB’s Mostofi Lab can determine that information with only an inexpensive WiFi transmitter and receiver.

You may be thinking they’re simply counting the WiFi signals given off by people’s smartphones, which is already an established technique. But, this is different, and works regardless of what the people have in their pockets. It measures the strength of the received signal that’s bouncing back from the crowded room. Each person causes a signal drop that can be registered, and their work has made it possible to count the individuals based on those signal drops. For now, it can only count a group of up to 20 people, but they were able to achieve 90% accuracy. Further development could potentially increase that number to cover much larger crowds.