University of Massachusetts Amherst

Search Google Appliance

Burleson Leads UMass Team Studying Transportation Security

As states move into the digital age and adopt “pay as you go” fee collection for toll roads and bridges, parking, and even public electric car recharging, user privacy, identity theft, and fraud are increasing concerns. Now a team of University of Massachusetts and Brown University researchers, led by Electrical and Computer Engineering Professor Wayne Burleson at UMass Amherst, has launched a three-year, $1.17 million study to identify the most efficient, low-cost, and reliable ways to provide secure, private, and trusted transactions.

The team of engineers, computer scientists, and transportation system researchers here, at UMass Dartmouth, and at Brown won funding from the National Science Foundation’s Trustworthy Computing program.

The study is entitled, “Pay-as-you-Go: Security and Privacy for Integrated Transportation Payment Systems (ITPS).” The team will investigate security and privacy for ITPS, which includes payments for trains, subways, buses, and ferries, plus tolls for roads, bridges, tunnels, parking, and recharging of electric cars.

As Burleson explains, “While a large body of research exists in the area of secure payments, the application of this knowledge to transportation systems is challenging because of security and privacy requirements and the need for a low-cost implementation. The unique aspect of this project is that we will explore deep problems in the disciplines of cryptography, secure systems, and transportation systems engineering motivated by and applied to a very pressing and tangible set of problems that impact almost all aspects of society.”

Others involved from the UMass Amherst campus are co-principal investigators John Collura, Civil and Environmental Engineering Department, Kevin Fu, Computer Science Department, and Christof Paar, Electrical and Computer Engineering Department. The project was initiated with funding from a UMass President’s Office Science and Technology Award, which included an ITPS workshop for academic researchers, government, and industry in Boston in February of 2009.

Public acceptance is central to successful designing and deploying new financing schemes, the researchers say. Experience has shown this is directly linked to consumers’ perceptions of system privacy, security, and safety.

Meanwhile, system operators strive not only to ensure secure and reliable collection, but to avoid compromising traveler mobility or environmental quality. To achieve this, system operators may need to observe certain aspects of user behavior, but consumers may object to such data being collected. Burleson’s team hopes to show that its approach can provide adequate privacy at a low cost while also allowing operators to collect the data necessary to improve their transportation systems.

UMass Amherst’s transportation engineer Collura says future uses of behavior data will allow public and private services to improve traffic management, travel time estimation, emergency management, congestion pricing, and air quality.

Payment devices for transportation systems must be extremely cheap, Burleson points out. They must be produced in mass quantities, often millions of tokens per year, but may only be used for a short time or even only once. This means that many sophisticated cryptographic techniques are much too computationally intensive to be used in such devices.

He adds, “Next-generation payment systems for transportation will support seamless travel by integrating various forms of transportation as well as parking and small-scale commerce such as buying gas or food, for example.”

Further, Collura says electronic payment systems already in use allow innovations such as dynamic or flexible pricing; for example, reduced fares for rides on overcrowded subways. Similarly, he says, although the idea is controversial, deploying a vehicle miles traveled (VMT) fee as an alternative to higher gas taxes as recently proposed in Massachusetts, Oregon, and other states, could be greatly facilitated by intelligent ITPS. (April 2010)