Sunday, August 26, 2007

Tracking containers.

The world is a very different place out beyond the horizon. Even as you read this, there are some 40,000 large cargo ships plying the world's waterways and oceans, not to mention innumerable smaller merchant craft, all pulling in and out of ports, loading, unloading, changing out crews and cargos, and steaming from one location to the next.

In what can be a very murky world of shadowy ship registry offices, lengthy manifests, and dockhands who change out faster than Barbosa's crew, how all these ships come by their cargo, how that cargo is loaded, by what polyglot seamen and in what untamed ports, can be an amazingly scrambled and trackless story rivaling the Pirates of the Caribbean.

Scenario: A single ship starts out in Singapore with containers filled with electronics, passes through Indonesia where it picks up spices, sails to Calcutta to load cotton, Port Said where it boards an Egyptian crew, Piraeus where it stops for fuel, Tangier where it picks up leathers, Scotland where it packs in woolen sweaters, and finally sets sail for Newark, New Jersey. Eleven million containers packed with such goods reach U.S. ports every year.

At any point in a merchant ship's journey, pry open container XYZ mid-ocean, and what might you find? When you can't be sure, that spells danger. The possibility that a single container has gone purposefully astray and might now be packed with explosives, or loaded with a virulent biologic destined for our shores, is not a fictional scenario. (In 1988, it was an Al Qaeda merchant ship that delivered the materials needed to bomb U.S. embassies in Tanzania and Kenya. That same ship was never seen again.)

Given lots of time, customs agents could find all contraband. But, in the world of maritime shipping, time is the enemy. Try delaying a delivery, and you may face some rough characters down at the docks (think On the Waterfront). What's more, anything that hinders the swift transit of goods around the world can have a rippling effect on the world's economy.

MATTS -- DHS S&T's Marine Asset Tag Tracking System -- is a miniature sensor, data logging computer, radio transceiver, and GPS tracking system integrated into a compact and inexpensive black box, about the size of a deck of cards. Affixed to a shipping container, MATTS can use its on-board GPS chip to estimate its location if the GPS signal is lost. And, in the final version of the system, containers outfitted with MATTS tags will be able to transmit through shipboard communications systems, even if they are placed deep below deck. The tag's signal will "jump" from container to container until it finds a path it can use. Better yet, this black box stores its location history and reports it back when in range (up to 1 km) of an Internet equipped ship, container terminal, or a cell phone tower. At any point in a container's journey, its history can be examined, and if anything has gone amiss, authorities know instantly to scrutinize that particular container.

Ultimately, MATTS will be integrated with S&T's Advanced Container Security Device. The ACSD sends an alert through MATTS when a container has been opened or tampered with on any side, providing even more security.

"MATTS will globally communicate in-transit alerts to Customs and Border Protection, and this capability satisfies a high-priority CBP requirement," says Bob Knetl, Program Manager for the MATTS research within S&T's Borders and Maritime Division.

In late April 2007, one hundred MATTS-equipped containers started out in the Port of Yokohama, Japan, and are now making their trans-Pacific crossing to the Port of Los Angeles/Long Beach, where they will then continue by rail to the Rochelle, Illinois, Rail Terminal and be unloaded and trucked to their final destination. This test, ending in August, will demonstrate that the communications can be used internationally (in this case, by Japan's Ministry of Land, Infrastructure and Transportation) and that transitioning to domestic drayage once portside in Long Beach also runs smoothly.

MATTS was developed under a DHS S&T Small Business Innovation Research (SBIR) contract by iControl Incorporated, a small Santa Clara, CA-based company.

"A serious threat is posed by the cargo that containers may hold," says Vinny Schaper, SBIR Program manager. "We have to view the ocean with grave concern, and realize that a maritime attack is not beyond the realm of possibility and if it comes, it will probably involve the use of merchant ships. Eleven million containers a year are brought onto our docks. Interrupt this with a terrorist attack, and the backup would reach around the world."

Note: This story has been adapted from a news release issued by US Department of Homeland Security.

Thursday, August 09, 2007

MIT wins Marconi Prize.

MIT Professor Ronald L. Rivest, who helped develop one of the world's most widely used Internet security systems, has been named the 2007 Marconi Fellow and prize-winner for his pioneering work in the field of cryptography, computer and network security.

Rivest, the Andrew and Erna Viterbi Professor in MIT's Department of Electrical Engineering and Computer Science, will receive the award and accompanying $100,000 prize at the annual Marconi Society Award Dinner on Sept. 28 at the Menlo Circus Club in Atherton, Calif.

The Marconi Society, established in 1975 by Gioia Marconi Braga, annually recognizes a living scientist who, like her father Guglielmo Marconi, the inventor of radio, shares the determination that advances in communications and information technology be directed to the social, economic and cultural improvement of all humanity.

"Ron Rivest's achievements have led to the ability of individuals across the planet--in large cities and in remote villages--to conduct and benefit from secure transactions on the Internet," said Robert Lucky, chairman of the nonprofit Marconi Society.

The group cited Rivest's advances in public-key cryptography, a technology that allows users to communicate securely and secretly over an insecure channel without having to agree upon a shared secret key beforehand.

"Public key cryptography has flattened the globe by enabling secure communication via e-mail, web browsers, secure shells, virtual private networks, mobile phones and other applications requiring the secure exchange of information," Lucky said.

A native of Niskayuna, N.Y., Rivest attended Yale University, where he earned a B.S. in mathematics in 1969. After receiving his Ph.D. in computer science from Stanford in 1974, Rivest accepted an offer to join the faculty at MIT.

At MIT he met two colleagues, Leonard Adleman and Adi Shamir, who would become his partners in solving the puzzle of public-key cryptography.

"Ron is a very special person," said Adleman. "He has a Renaissance quality. If tomorrow he discovered an interest in rocketry, then in five years he would be one of the top rocket scientists in the world."

Fortunately, what captured Rivest's imagination was the challenge of a public key encryption system. He managed to enlist Adleman and Shamir in his quest to produce what he called an "e-crypto system." It was a challenge ideally suited to Rivest's mathematical interests, relying on what Adleman calls "some of the oldest and deepest mathematics, going back to antiquity."

In public key cryptography, there are two keys; one known to everyone, and one known only to the recipient. The public and private keys are paired in such a way that only the public key can be used to encrypt messages and only the corresponding private key can be used to decrypt them. But even if someone knows the public key, it is effectively impossible to deduce the private key. To design such a system was the challenge. In effect, it was a mathematical puzzle.

The RSA encryption algorithm that Rivest, Shamir and Adleman developed relies on the challenge of factoring large numbers (typically 250 or more digits long), a problem that has stumped the world's most prominent mathematicians and computer scientists for centuries.

At one end of the "conversation," the receiving party's computer secretly selects two prime numbers and multiplies them to create a "public key" which can be posted on the Internet. On the other end, the sending party's computer can take that key, enter it into the RSA algorithm and encrypt a message.

The genius of the scheme is that only the recipient knows the prime factors that went into the creation of the public key--and that is what is required by the RSA algorithm to decipher the message. Even though others can see the encrypted message and the public key, they cannot decipher the message because it is impossible to factor the number being used in the public key within a reasonable period of time.

The team developed its system in 1977 and founded RSA Data Security in 1983. RSA was acquired in 1996 by Security Dynamics, which in turn was acquired by EMC in 2006. Rivest has continued his work in encryption and is the inventor of the symmetric key encryption algorithms RC2, RC4, RC5, and co-inventor of RC6.

looking at adds

How do consumers look at advertisements? Most marketing textbooks advance the theory that looking at ads is a predominantly "dumb process," driven by visual stimuli such as the size of the ad or the color of the text.

However, new research by researchers from the Netherlands and the University of Michigan uses eye-tracking software to reveal that it may be our goals -- the tasks we have in mind -- that drive what we pay attention to, even during a few seconds of ad exposure.

In the August issue of the Journal of Consumer Research, Rik Pieters (Tilburg University, The Netherlands) and Michel Wedel (University of Michigan) perform an eye tracking experiment on 220 consumers. The consumers are split into four groups, each with a different goal, and given free rein to view a series of advertisements.

The study is self-paced -- that is participants are allowed to look at the ads for as long or as short of a time as they would like. Overall, the participants looked at the 17 target ads in the study for an average of about 4 seconds only -- but with notable differences in focus.

Those asked to memorize the ad focused on both the body text and the pictorial representation of the product. Those asked to learn about the brand, on the other hand, paid enhanced attention to the body text but simultaneously ignored the pictorial.

This supports the Yarbus thesis that ad informativeness is goal-contingent. Differences in pupil diameter between ad objects but not between processing goals reflect the pupil's role in maintaining optimal vision.

"The fact that even during the few seconds of self-paced ad exposure, attention patterns already differ markedly between consumers with different goals underlines the importance of controlling and knowing consumers' processing goals in theory building and during advertising pre- and post-testing," the researchers write.

In other words, the eyes are a reflection of consumer goals.

Reference: Rik Pieters and Michel Wedel. "Goal Control of Attention to Advertising: The Yarbus Implication," Journal of Consumer Research: August 2007.