Video Player

Network-Driven Transportation

Li-Shiuan Peh: November 3, 2009; Running Time: 57:15

About the Lecture

About the Lecture

Today, cell phones are a menace to safe driving, as they distract operators who should otherwise focus on the road. Tomorrow, cell phones could actually improve our driving, and help drivers avoid traffic congestion, use the road system more effectively, and manage the parking supply. Li-Shiuan Peh says that the key to these services are future mobile devices that will have the computer power equivalent to today’s large servers in data centers. Combined with rapid advances in wireless networking, these mobile devices will be harnessed to provide new apps, like next generation transportation programs.

We currently use the Internet and Wi-Fi or 3G and then run our programs in the cloud on heavyweight servers. Peh says that an opposite case is likely to emerge, with a move towards collaborative computing, using mobile devices and localized cell phones to replace the heavyweight servers. She envisions a time when advanced cell phones will be “stitched together” to run a single piece or information or a program. Peh says this grassroots type of computing will appeal to the general public, “the sociology” of users, who like to be involved in transportation activities. Behind this collaborative computing, engineers are fine-tuning a sophisticated mesh-network of communications.

One of the key protocols for the mesh network is for dedicated short-range communications (DSRC). It is vital for mobile applications, like accident prevention, because it is micro-seconds faster than current standards. DSRC will have very high local coverage, provide faster and more complete transmissions than existing cell towers, and, in particular, be able to overcome the coverage issues of tunnels and dead-spots. Moore’s Law scaling would predict that the computing power needed to advance DSRC applications would be more powerful and more efficient that what we know today.

Location resident services are quite demanding from a communications point of view. Engineers must design around the fact that the phones must be able to hand-off information and exchange it with each other, since the handsets/cars will move in and out of a cordon. Peh notes that existing intelligent vehicle programs in the United States, Europe, and Singapore are using elements of mesh communications today, and that collaborative computing will become the better solution. Mesh computing can also have very local and practical applications. Stitching cell phones, cameras, and databases together could provide a real-time tool for Amber Alerts, and help solve problems that require geographic sensing, such as locating missing children. Similar sets of technologies could be stitched together to reduce congestion and fuel use as they guide drivers to open parking spots.

Lecture Details

Location: 32-124

“Since all of these architectures will all use meshes of some sort to communicate between devices and cars, so why don’t we use them for computing as well? As we add multiple hops… perhaps we could actually run some application software on our individual cell phones, on our neighboring cell phones and use that as a single big distributor and parallel computer. ”

Li-Shiuan Peh

About the Speaker

Li-Shiuan Peh

Associate Professor of Electrical Engineering and Computer Science

Li-Shiuan Peh’s research interests include interconnection networks (networks connecting subsystems within a digital system, such as multiprocessors, blades, disks, clusters, router line cards, on-chip modules, embedded systems, etc); more specifically power-aware interconnection networks; parallel architectures and networking in general.

She received a B.S. from the National University of Singapore in 1995 and Ph.D. from Stanford University in 2001.

Li-Shiuan Peh

Peh home page

About the Host

About the Host

Transportation@MIT

Transportation@MIT is a coordinated effort to address one of civilization’s most pressing challenges: the environmental impact of the world’s ever-increasing demand for transportation.

Building on MIT’s rich tradition of engineering research and interdisciplinary collaboration, the initiative knits together the wide-ranging, robust research already under way at the Institute and creates new opportunities for education and innovation.