Internet gurus aim for the stars
May 24, 2001
E.T. may not be able to phone home anytime soon, but the lovable alien may be able to send e-mail--if a draft proposal released this week for an interplanetary Internet takes flight.
Described as a "work in progress," the proposal to the Internet Engineering Task Force--the group that sets standards for the Net--calls for terrestrial testing of interplanetary Internet protocols later this year, with a live test onboard the NASA Mars mission in 2003.
"What we are fundamentally about is deploying as much re-usable, standardized communications infrastructure as we can afford around the solar system, so that future missions don't have to take it all along with them," said Adrian J. Hooke, manager of the DARPA InterPlaNetary Internet (IPN) Project and co-author of the proposal. "They can use capabilities put in place by other missions."
The proposal calls for a network of Internets to facilitate communication among planets, satellites, asteroids, robotic spacecraft and crewed vehicles and create a stable interplanetary backbone network.
While it may sound like a pie-in-the-sky plan, the person heading up the project is none other than Vinton Cerf, an MCI WorldCom vice president and creator of the TCP/IP standard--the foundation of communications on the Internet.
Furthermore, the project has drawn the interest--and funding--of both NASA and the Defense Advanced Research Projects Agency, which could use it as a blueprint for future work.
Others working on the project include members of the Jet Propulsion Laboratory, Mitre, Global Science and Technology, and SPARTA. Other parts of NASA have already begun experimenting with remote Internet communication, demonstrating last year that it could use standard Internet protocols to communicate with an orbiting spacecraft just like any other node on the Internet.
Planetary packet loss?
Like the Pony Express, such a "network of Internets" would have to deal with the lengthy delays that seem inevitable when communicating across the vast distances of space, Hooke said.
"It's 'e-mail-like' in that as the propagation delay increases it gets increasingly clunky to try to communicate in a chatty mode," he said.
When the objects are at their closest approach to Earth, light takes a little more than a second to reach the moon, almost 6 minutes to reach Mars and 9 minutes to reach the sun--not to mention the 4 years it takes light to reach the nearest other star, Alpha Centauri.
"Store-and-forward" networks would connect the more localized Internets, Hooke said. "Anywhere that there is a short-delay environment--around Earth, within a free-flying spacecraft, on and around another planet--we can use communications protocols that are fairly closely related to their terrestrial Internet counterparts."
Moreover, the organization of the network can change. Today, the quickest way to send a message to Mars would be directly, when Earth and Mars are close. Yet when the two planets are on opposite sides of the Sun, routing data through a third node--say at Mercury or Venus--would make the transmission more reliable.
Space poses other problems as well. Signals are more easily intercepted by others, making sturdy security a must, and data loss will be a routine occurrence, warns the draft.
On the Internet, the transfer of data from server to server is accompanied by acknowledgements and error messages in real time. With long delays plaguing the communications, however, such messaging would just slow things down, stated the proposal.
To handle these considerations, Cerf and the others propose a technology called "bundling" to connect Internets together. Such bundles of data will minimize the traffic between interplanetary Internet servers.
Space's common tongue
Furthermore, a standards-based approach to interplanetary communications will allow NASA to reuse communications data from mission to mission.
The use of the technology is not that far away, either, supporters contend.
Mars would be the first addition to the interplanetary Internet. By 2010, the group believes as many as seven satellites dedicated to communications could be in orbit around the planet.
Other missions could then benefit from the communications hardware on those satellites. "Doing this right, and in a way that can be shared among many projects and missions, makes a lot of sense," said Jerry Fiddler, chairman and co-founder of embedded software maker Wind River Systems. The company's operating system and software has been included in such missions as the Mars Pathfinder, the Clementine mission to the moon and the International Space Station.
However, the proposal states that the ultimate test will be commercial applications: "While such developments may still lie decades in the future, the potential investment and benefits can be appreciated as we contemplate the explosion of new markets associated with the commercialization of the Internet."
When that happens, the draft's authors state, the interplanetary Internet should really take off.
By Robert Lemos, Special to CNET News.com, Copyright ©l 1995-2001 CNET Networks, Inc.,