acm-header
Sign In

Communications of the ACM

ACM TechNews

Bluetooth and Wi-Fi Sensing From Mobile Devices May Help Improve Bus Service


View as: Print Mobile App Share:
University of Washington researchers last spring tested an inexpensive system to sense Wi-Fi and Bluetooth signals from bus passengers mobile devices on the university's shuttle buses.

Transportation engineers at the University of Washington have developed an inexpensive system that uses Wi-Fi and Bluetooth signals from passengers mobile phones and devices to collect better data about where bus riders get on and off, how many people us

Credit: University of Washington

University of Washington (UW) researchers say they have developed a low-cost way to use Wi-Fi and Bluetooth signals from bus passengers' mobile phones and devices to harness useful data.

The researchers developed sensors costing about $60 for each bus, which detected a unique identifier called a Media Access Control (MAC) address. The system only collected MAC addresses and the time and location they were detected from Bluetooth or Wi-Fi signals, and each address was anonymized for privacy protection.

The sensors mounted inside the buses initially picked up more than 20,000 unique addresses from mobile devices, so a key challenge was developing processing algorithms to filter out signals from people who were near the bus but not actually riding it. "That's probably the hardest part of the whole thing," says UW researcher Kristian Henrickson.

The researchers eliminated signals that were unreasonably long or short, or that emerged or vanished far from a bus stop, resulting in 2,800 reliable "trips."

"Let's say you have a Husky game or Seahawks game and you want to know how much demand changes so you can offer the right level of bus service for this special event," says study senior author Yinhai Wang. "If you can gather enough data from these real-time sensing systems, that's going to offer very valuable information."

From University of Washington News and Information
View Full Article

 

Abstracts Copyright © 2016 Information Inc., Bethesda, Maryland, USA


 

No entries found

Sign In for Full Access
» Forgot Password? » Create an ACM Web Account