Although the "Interplanetary Internet's" ultimate focus is shuttling information throughout the solar system, the implications here on Earth for using its new class of network, the Delay Tolerant Network (DTN), could mean improved communications for the military and for emergency crews.

While the worldwide Web community is focused on the Earth-bound issues of bandwidth, convergence, Internet2, and the Next Generation Internet, a considerable effort is afoot to leapfrog this terrestrial Internet in favor of a deep-space network of Internets. The project is called the Interplanetary Internet. A MITRE research team is at the heart of the development process, working together with The Defense Advanced Research Projects Agency (DARPA), NASA's Jet Propulsion Laboratory, Global Science and Technology, and SPARTA, Inc., with oversight from Vinton Cerf, originator of the terrestrial Internet's key networking protocol TCP/IP.

Time and distance represent the biggest challenges to any information network attempting to span the cosmos. On Earth, information traveling at the speed of light makes both wired and wireless transmissions near instantaneous from sender to recipient. Not so in outer space where distances in the millions of miles can make light speed take up to 6 hours to reach certain planets. Compounding such transmission delays is the fact that, unlike the stationary terrestrial Internet, planets move. Mars, depending on its orbit around the Sun, can be anywhere from 35 million to 245 million miles from Earth. That's a time difference between 3 and 20 minutes—an intolerably unworkable time delay for any Internet.

Obviously then, a new way to relay space-borne digital information is necessary, which is exactly what MITRE's Bob Durst and Keith Scott have been working on for over two years. "When we send messages on the Earth Internet, there is typically an assumption of essentially instantaneous connectivity. As a result, communications processes on remote computer systems tend to be highly conversational," explains Durst. "To speed up the process and make the best use of resources, each message is sent in a train of digital packets, with each packet making up only a portion of the entire message. The parts of the message are reunited by computer on the way to its destination. But this approach won't do for deep space. A single packet with its message and all supporting information totally self-contained will be necessary." Limited opportunities for communication and short communication contact times make interactivity between sender and recipient impractical and inefficient. Durst and his mates on the Interplanetary Internet Core Team had to rethink the terrestrial model completely. They came up with the Delay Tolerant Network (DTN) and its single, self-contained information vehicle, the bundle.

A bundle, much like a single package from a parcel post delivery, would be less affected by the delays and disconnectedness characteristic of delivering messages in the vastness of outer space. Theoretically, a bundle would be delivered to a gateway here on Earth, which would decide on the best route to a given destination, taking into account distance to the destination, bandwidth and memory needs, transfer delays, expected episodes of disconnect—as in waiting for a connection to a Martian satellite as it orbits into proper position—as well as the availability of further relay sites to pass the bundle along over the entire delivery route.

Contacting Mars

NASA has three such potential gateways in its Deep Space Network, one each located in California, Australia, and Spain. Each facility consists of a set of giant antenna arrays—the largest some 70 meters in diameter. And just like postal service deliveries, each bundle will have its own receipt and tracking number with which one can follow the delivery from relay satellite or transiting spacecraft to the next relay point and so on to its eventual destination. That's a big challenge, but headway has already been made.

According to Durst, the concept for this Delay Tolerant Network is in place now. And the team has "some decent code to release to the public." Like the path to accelerated success taken by Linus Torvald with his Linux code, this code is open source, which allows experts from around the world to tweak the code, work out bugs, and suggest improvements. When the network is completed, it must be tested. For that endeavor the Red Planet is the candidate. It's the nearest planet to Earth and NASA's activity involving Mars affords a ready opportunity to piggyback on each mission, small relay satellites called "marsats." Eventually, with a constellation of marsats orbiting above the Red Planet, the array would form the backbone to a Martian Internet.

The Martian Internet would immediately benefit all communications to and from Mars. Consider, for instance, the Mars Pathfinder of 1997, which could send only 30 megabits of data back to Earth each day, which averages out to 300 bits per second. In comparison, the average home Internet modem of today, a 56kb modem, transmits data at the rate of 56,000 bits per second. A Martian network would more than treble the Pathfinder's transfer rate, but even that's not peppy enough to transmit visuals, especially video. However, a satellite in deep orbit to Mars, say, 10,000 miles above the planet, could crank out 1 megabyte of data—over 8 million bits per second—or the equivalent of a continuous video feed. Such a data rate would hasten NASA's exploration and understanding of the planet. Then too, perhaps the ill-fated Mars Polar Lander, which crashed on landing, could have used such a Martian Internet to communicate its approach while descending.

Advantage Home

According to Adrian Hooke, manager of NASA's space missions' standardization programs, the benefits of Delay Tolerant Networks also present considerable advantage for terrestrial communications. Battlefield communications among "untethered nodes" (helicopters, aircraft, and satellites, which experience episodes of delay because of distance, terrain, or even radio silence) would gain from such a network. Extending the concept to civilian uses, Hooke sees advantages to "intelligent" highway systems and even emergency crews working in hostile environments.

Although no one should start printing Yellow Pages for Mars quite yet, Durst is confident that a semblance of a Martian prototype network, albeit on a small scale for now, will be in place relatively soon with a more full-blown network to follow—but much more slowly. Durst says that funds for interplanetary exploration are tight and launch opportunities are scarce, which means less chance to piggyback marsats for the Martian Internet. Hooke predicts that 2010 would be an appropriate timeframe for a robust Martian Internet. If so, ET will have to wait a while longer before clicking home.

In the Year 2015

A geologist in his earthbound laboratory closely examines a Martian rock—a virtual Martian rock, that is—transported to him from some 35-million miles away courtesy of a soon-to-be-developed, space-borne Internet. He chips a minute piece from the rock, dissolves it in an acid bath and then runs both optical and chemical analyses. Soon test results begin filling up his computer screen.

Back on the Red Planet, a robotic Martian laboratory gently cradles the rock, quickly obeying every test command while relaying constant 3-D imagery and test data to a small constellation of orbiting mini-satellites. The satellites, in turn, transmit the information to a deep-space satellite and then on again to a gateway on Earth, where it is downloaded to the geologist. Although such a scenario is still out of reach for the moment, the plan for shrinking a bit of the final frontier is well underway.

—by Tom Green

Related Information

Websites

• National Aeronautics and Space Administration (NASA)

Solutions That Make a Difference.®
Copyright © 1997-2013, The MITRE Corporation. All rights reserved.
MITRE is a registered trademark of The MITRE Corporation.
Material on this site may be copied and distributed with permission only.

Privacy Policy | Contact Us