Yingying Chen, Stevens Institute of Technology: Detect Driver Cell Phone to Decrease Accidents

December 9, 2011

To increase safety on the road, wireless researchers are looking for technologies that can passively influence the behavior of drivers who put themselves and others at risk through distracting cell phone use. For their paper demonstrating a technique that detects drivers using phones, Dr. Yingying Chen and her graduate students in the Department of Electrical and Computer Engineering at Stevens Institute of Technology recently won the Best Paper Award at the ACM International Conference on Mobile Computing and Networking (MobiCom 2011).

"Their research addresses the problem of distinguishing between a driver and passenger using a mobile phone, which is a key milestone for enabling numerous driver safety and phone interface enhancements," says Dr. Yu-Dong Yao, Department Director for Electrical and Computer Engineering. "MobiCom is a competitive and highly-regarded conference for mobile computing and network research, and it is a great honor to receive the Best Paper Award."

The research was led by Dr. Chen at Stevens and Drs. Marco Gruteser and Richard Martin of WINLAB at Rutgers University. The other authors of the paper, "Detecting Driver Phone Use Leveraging Car Speakers," are Jie Yang, Simon Sidhom, Gayathri Chandrasekharan, Tam Vu, Nicolae Cecan, and Hongbo Liu.

The team developed a cell phone detection scheme using an acoustic approach wherein a phone leverages the built-in Bluetooth and car stereo to generate a series of high frequency beeps over the stereo. The phone records these beeps, which are spaced in time across the left, right, and if available, front and rear speakers, and times their arrival. Using a differential ranging approach to estimate the phone's distance from the car's center, a passenger or driver classification can be made.

Although a car is a relatively small space, a moving car with the radio on or windows down is an extremely complex acoustic environment. Despite this multipath noise, experimental trials with the team's detection technique, using two different phones in two different cars, demonstrated that the customized beeps are imperceptible to most drivers yet robust to background noise. They achieved accuracy in pinpointing a driver's phone up to 95% of the time, with a low false positive rate.

Their passive technology approach reflects statistics that demonstrate hands-free phone operation has not effectively reduced phone-related accidents. In 2009 alone, the National Highway Traffic Safety Administration identified cell phone distraction as a factor in crashes that led to 995 fatalities and 24,000 injuries. And while simply conducting a cell phone conversation provides more than enough distraction to increase a driver's accident risk, the wide range of activities available on mobile devices—including texting, checking email, using navigation, and even playing games while driving—has the potential to create extremely dangerous situations.

Mitigating dangerous driver behavior requires a detection technique such as proposed and tested by Dr. Chen and her colleagues. Their demonstrated success provides a foundation for future applications that will assist drivers to maintain safer cell phone habits.

As an Assistant Professor at Stevens, Dr. Chen leads the Data Analysis and Information Security Laboratory (DAISY Lab), which facilitates research into access and security of data on wireless networks. She has been the PI on multiple National Science Foundation (NSF) grants as well as the recipient of the prestigious NSF CAREER Award. Dr. Chen's research addresses wireless security and the challenges faced as wireless networks become increasingly pervasive and ubiquitous. Her work supports new approaches to conventional and wireless security by building location-oriented information into the wireless stack.

MobiCom 2011 was the 17th annual iteration of the conference dedicated to addressing challenges in mobile computing and wireless and mobile networking. A premiere international forum on the systems and applications that progress our mobile computing power, MobiCom covers a wide range of topics and is highly selective.