7 min read
NASA
Next in trending

We’re building an interplanetary internet

Prepare yourself for cosmic lag

We’re building an interplanetary internet

Prepare yourself for cosmic lag


The very first conception of what would eventually become the internet may have been a series of memos written by an MIT computer scientist named J.C.R. Licklider in August 1962. In them, he described a “galactic network”, where anyone could access data and programs from any location in the universe.

Licklider went on to become the first head of the computer research program at DARPA, where he convinced his successors of the importance of networking. The result of their work is what you’re staring at right now — a network where anyone can access data and programs from any location in the world.

But so far the internet has been mostly limited to the surface of Earth. Realising Licklider’s vision of expanding that network into the heavens is difficult for several reasons.

Laaaaaaag

The first is lag. On Earth, the distances we communicate over are pretty small, galactically speaking. Our signals travel at the speed of light, meaning they can get from Britain to Australia in 50 milliseconds. But to get to our nearest planetary neighbour, Mars, it takes between three and twenty-two minutes depending on the orbits of both planets.

That’s not too bad if you’re a research scientist working on a rover. But if you’re trying to play Counterstrike with some Martians, you’re going to have significant difficulties. In fact, the way that the internet is currently built — over TCP links — won’t really function at all with those delays.

Instead, researchers are working on different kinds of delay-tolerant networks. Traditional networks work by establishing a route and then sending the data down it. Delay-tolerant networks take a more forgiving approach, storing data at points along the way and then sending it onward when a connection can be established to the next node.

That means slow transmission, of course, as satellites go around the backsides of planets and the pesky Sun gets in the way. But the information does end up intact at its destination, eventually.

If all that is tough to imagine, picture a football game. Right now, we’re trying to score goals by hoofing the ball from our own goal line. That works when there aren’t any other players on the pitch. But when there are, it’s more useful to pass it to a defender, who passes it to a midfielder, who crosses it into the box where it’s headed home.

Few beta testers

The second problem is that we don’t have anyone on the other end to receive the data yet. The furthest permanent outpost humanity has managed is the International Space Station in Earth’s orbit, which is connected to the regular internet.

Artist’s impression of a Martian colony // NASA

The next stop could be Mars, but it’ll more likely be the Moon or one of the Lagrange points in between. Meanwhile, the robots that we’ve currently got stationed around various planets in the solar system and beyond can help us test the protocols, but they’ll make about as much attempt to understand any broken communications as a Frenchman would.

Regardless, without some humans (or sentient creatures of some sort) on the other end, an interplanetary internet is only going to be of limited use to mankind.

Timekeeping

Another challenge for network nodes stationed off-world will be timekeeping. Most space missions need some form of clock to know when to communicate with Earth. When an Earth-link can’t be set up, the various nodes of the interplanetary internet will need some way of synchronising their time.

Right now, all spacecraft clocks are coordinated from Earth using the Deep Space Network’s three overlapping antennas in California, Spain and Australia, though this system is starting to feel the stress of having to cope with missions launched as far back as 1977 while its telescopes are losing funding.

An interplanetary internet could be used to relieve the stress on that network by allowing our distant spacecraft to get some information, like time signals, off their neighbours rather than having to constantly call back to Earth. As we begin to look more closely at the outer reaches of the solar system, that’s going to be very useful indeed.


It’s already happening

Now here’s the exciting bit. We actually have the beginnings of an interplanetary internet already set up, and the man leading the charge is none other than one of the fathers of the terrestrial internet — Vint Cerf.


Cerf, in his role as Google’s chief internet evangelist, is working with NASA to develop the delay-tolerant networks that are required for deep space communications, and test them with existing spacecraft.

His team has already successfully used these protocols to transmit dozens of space images to and from NASA’s Epoxi spacecraft, which is en route to a comet and currently about 20,000,000 miles from Earth, without issues.

Cerf’s protocols have also been used to control a rover on Earth from the International Space Station. Astronaut Sunita Williams was able to control a Lego vehicle in Darmstadt in Germany from orbit, just like an astronaut may want to from a capsule orbiting another planet.

More recently, in September 2013, NASA used lasers mounted on a lunar craft called LADEE to test optical transmission of data through space. That can be done far, far faster than the Deep Space Network’s communications protocols, which have speeds comparable to a 28.8kbps modem from 1996.

However, just like we were able to use the internet in 1996, we’ve been able to get a version of an interplanetary internet working over those connections.

MarsNet

In Cerf’s talk at TEDxMidAtlantic in 2011, he tells the story of how NASA’s 2004 Martian rovers, Opportunity and Spirit, were originally intended to transmit directly back to Earth from the surface of Mars at 1996 modem speeds. However, when the radios were turned on, they overheated so they had to be used less intensely — reducing speeds further.

That’s when someone noticed that the rover also had an X-band radio, and that a similar radio was present on Martian orbiting spacecraft that had been used to map the surface in preparation for the rovers’ arrival and was no longer being used.

The engineers reprogrammed both the rovers and the orbiter to accept data from the rover and then re-transmit it back to Earth in a system known as Electra. This increased transfer speeds four times to 128kbps — about the same as ISDN speeds from 1999 — thanks to the orbiter’s position outside of the Martian atmosphere. In the process, the first interplanetary network was created.

Electra // NASA

When the Phoenix lander arrived at Mars’ north polar region in 2008, it was unable to contact Earth directly in any configuration, so it joined this network to send data back to Earth. Curiosity followed suit in 2011, taking the total of devices on the Martian end of the network to seven, including three orbiting spacecraft — Mars Global Surveyor, Odyssey, and Mars Express. Contact with two of the nodes has been lost, however — Spirit and Mars Global Surveyor are no longer operational.


Today, speeds on the Electra network have been boosted to more than 1,000 kbps, and work continues on the protocols. Unfortunately, the field was dealt a setback by the cancellation in 2005 of the Mars Telecommunications Orbiter, which would have allowed for laser communication between the Earth and Mars.

The aforementioned LADEE lunar spacecraft is now testing those communication principles instead, albeit over a much shorter distance, and it’ll get put through its paces again in 2017 when the Laser Communications Relay Demonstration mission is launched on a commercial satellite.

Meanwhile, NASA’s Mars network will be augmented by its recently-launched MAVEN spacecraft when that arrives in September 2014, though India’s Mars probe MOM, which launched a few days before MAVEN, will communicate directly with Earth.

Cerf has bigger ambitions, though. Towards the end of his TEDx talk he envisions our interplanetary network being connected by femtosecond laser pulses to the next nearest star to Earth — Alpha Centauri, about 4.4 light years away.

To detect those signals, we’ll need a network of receiver stations spread out across the entire solar system to account for the beam widening of the laser.

Sounds like the perfect test case for our brand new interplanetary internet.


Looking Up is a collection on Medium that offers a home for those obsessed with the world above our heads. It’s curated by @duncangeere. If you enjoyed this article, please click the “recommend” button below, and if you want more, then click the “follow” button to make sure you don’t miss anything we publish in the future.