The Internet In Space - What Is Happening?

Proposed Implementation
Current Status
The Future


In 1973, Dr. Vint Cerf co-wrote a pioneering white paper on the Transmission Control Protocol (TCP), the protocol that spawned modern Internet communication. It took 20 years for the Internet to take off and Cerf is now preparing for future communication needs.

A project he is currently working on is taking the Internet into outer space. The concept is not a new one. Fifty years ago, countries started sending probes into space to explore the solar system. It became important to communicate with these spacecraft using bi-directional radio links so as to be able to send instructions, and for the probe to send data back to earth. This was done using large Earth-based antennas as gateways, such as the Deep Space Network (DSN) created by the National Aeronautics and Space Administration (NASA). Initially, the protocols used were unique for each mission but as more and more mission were launched, all sharing the same ground infrastructure, people began to start standardizing the way in which the information was exchanged.
The idea of the Internet in space is an extension of this standardization. The DSN uses three research facilities strategically placed around the world to constantly observe spacecraft as the Earth rotates. It is managed and operated for NASA by the Jet Propulsion Laboratory (JPL). Within the JPL, the InterPlanetary Network (IPN) manages the program.
Tied in with this is the Mars Network, which is also being studied at JPL as a possible future element of NASA's Mars Surveyor Program. Mars is the nearest planet to us in distance and environment. The Mars Surveyor Program is designed to support Mars global reconnaissance, surface exploration, sample return missions, robotic outposts, and eventual human exploration. To facilitate this, the Mars Network wants to develop a communications system that will provide an increase in data rates and connectivity between Mars and Earth and to develop a navigational system on Mars that will allow more precise locational information while approaching and on Mars - essentially building a publicly accessible gateway on Mars. The first step for the IPN will be to use the proposed gateway on Mars and link it to Earth. To do this, there are many challenges to overcome, which will be discussed later. The current TCP/IP protocol can be used on either planet for the local Internet. Once the problems are solved and the astronauts can email home, it will be time to link the other bodies in the solar system to the network.

Proposed Implementation

The astronomical task that is the Interplanetary Internet is difficult to visualize. We find ourselves asking - how on earth is it going to work?

Imagine the Internet on Earth and a similar one on Mars, linked by satellites, which relay the information between the Internets. These special purpose satellites are called 'gateways'. Now think of it not only as two Internets, but a network of Internets, interconnected by a store-and-forward "overlay" network that forms the backbone across interplanetary space. Each Internet will have a local gateway.

To understand gateways further they could be compared to a local Internet Service Provider (ISP), which allows you to get access - via phone, Digital Subscriber Line etc. - to the terrestrial fiber backbone. The interplanetary gateways will be found either in a ground station on a planet's surface or on a dedicated spacecraft in orbit, and will act as data processing stations. On one side, the gateway will be connected to a local Internet in the usual way, on the other side to the interplanetary backbone via a radio or optical communications transceiver.

Each Internet will have its own protocol. Communication between protocols is complicated, but manageable. Every Internet protocol will be terminated at a local gateway. Then there is the problem of communicating from gateway to gateway, this is overcome by a new "long haul transport" protocol which will communicate between gateways. A new, end-to-end "bundle" protocol will operate above the transport layer to carry information from a gateway on Earth to (for example) one on Mars. A 'bundle protocol' is needed because the IPN will not be able to communicate in real-time due to the distances that data will have to travel, so instead of sending small packets of data piecemeal, the information is gathered up into a bundle at the interplanetary gateway. When the bundle is complete, it is sent off in one big burst to the next gateway.

For the home user, satellite Internet access is a lot like satellite television. A satellite orbiting the Earth beams data to a dish attached to your house. Instead of sending information from the dish to the television screen, as in satellite television, the dish relays the data at speeds of 400 kbps or more to a special satellite modem connected to your PC.

Current Status

The JPL has performed a conceptual study of a the proposed Mars Network, focusing on detailed design of a prototype small satellite known as a Microsat. The Mars Network would consist of many Microsats for communication around Mars, and at least one larger, more capable MARSat satellite acting as a higher bandwidth gateway to Earth.

This is in keeping with the overall plan of the IPN - to create remote Internets elsewhere in the Solar System linked by an Interplanetary Backbone Network (IBN). The remote Internets can use protocols similar to Earth's Internet, whereas the IBN requires a major new development effort. Much work done in creating wireless networks on Earth will be applicable to remote Internets, where ground-based networks are impossible.

The IPN will vastly simplify the communications element of future space exploration missions. Currently, each mission must incorporate it's own tailored communications infrastructure. In future, missions will simply use the IPN.

The JPL has also examined how, when, and in what order and configuration to launch such satellites, in order to have a Mars Network up and running as soon as is possible.

Finally, the JPL has explored the creation of a Mars Exploration Gateway to link the Mars Network and the Internet. In partnership with academia and industry, NASA has plans to make this gateway accessible to the public as much as is possible. Public involvement in future space exploration missions would increase NASA's public profile and strengthen its case for public funding. However, this capability will demand security of the highest order.

A working prototype called the CCSDS File Delivery Protocol (CFDP) has been developed and experimental implementations of it exist. It is extremely limited in scope and was designed to support individual space exploration missions but served as a basis for further work.

The IPN Research Group, headed by Dr. Cerf, has published an Architectural Definition of a possible protocol for the IPN as a working document of the Internet Engineering Task Force (IETF). It proposes a store-and-forward overlay network above the transport layers of the underlying networks composing the IPN. The protocol is designed to operate in environments that have very long speed-of-light delays and across intermittently connected networks. Security, naming and routing in the IPN are discussed.


The increase in the number of Mars exploration elements and Mars Networks brings with it an increase in the complexities involved trying to coordinate the communications occurring between them. In an attempt to solve this problem, Mars Network (which is being studied at JPL as a possible future element of NASA's Mars Surveyor Program) is working with the Internet community to design a Mars Internet. They attempt to do this using an IP-like protocol that facilitates file-level communications between exploration and constellation elements and is robust enough to contend with long round-trip light-times and noisy, intermittent, power-constrained deep space links.

At the moment, communications between spacecraft and Earth are intolerable. Data is transmitted between space and Earth via antenna clusters on three continents, placed approximately 120 degrees apart around the world. Problems arising between communications are a result of an inadequate bandwidth. With more than 40 active missions a month vying for time on the network, the system is becoming increasingly overloaded.

Robotic probes deployed by NASA must ensure that reliable communication links exist between these themselves and Earth. Such reliable links are not easily achieved, since communications performance decreases as the square of the distance increases. A satellite transmitting back 400 million km to Earth from Mars would make communications 100,000,000 times more difficult. This astronomical distance between planets also results in long delays between parties or subsystems e.g. a round-trip transmission from Earth to Mars could exceed 40 minutes.

Security is also an issue. Currently, Internet security protocols require a series of 'handshakes' to certify correct identity and permissions etc. This approach will not work for the IPN, as real-time communication is impractical. Therefore, we need to incorporate SSL (Secure Sockets Layer) and other protocols currently being used for e-mail security.

Antennas need to know where something will be and when it will be there so as to be ready to send or receive a signal. They also need to take into account celestial mechanics and the fact that transit times change as things orbit the sun. There is also the annoying fact that the sun can block signals for weeks on end when two sources find themselves on opposite sides of the solar system.

The Future

The Interplanetary Internet will expand slowly, one component at a time. The primary means by which the first components of the IPN are to be put in position is by ‘hitching rides’ on spacecraft built for other projects. An example of this is the Mars Odyssey, an orbiter which was launched in April 2001, carrying a telecommunications relay package that is intended as the first relay for future landers, such as robotic surface explorers. Satellites, which generate more power than surface landers, could then amplify the landers signal. This ultimately means more data with fewer errors.

Sending this telecommunications orbiter is the first step towards the plan of using a stationary spacecraft that would hover above a lander just like an Earth geostationary satellite. When this idea becomes a reality, believed to be possible by 2010, it means that landers on the surface of a planet will be continuously connected at very high data rates to its orbiting relay.

However, it is not only the US who are launching Mars bound orbiters, the European and Japanese space agencies will each have an orbiter arriving there in the 2003 time frame.

As each new satellite is launched, the Interplanetary Internet gets both more resilient and more powerful. The additional devices give the data more possible routes back to Earth, so if one fails, contact is still possible. For example, today the quickest way to send a message to Mars would be to transmit it directly. 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.

In 2007, NASA will collaborate with the Italian space agency to send an orbiter to Mars. This will be the first orbiter dedicated to having telecommunications as its primary function. It will have capabilities of up to 1Mbps continuous data rate from the surface of the planet back to Earth, or about 85 gigabits of data per Martian day.

According to Vinton G. Cerf, MCI WorldCom’s senior vice president, by 2008 there could be as many as seven communications and navigation satellites in LMO (low Mars orbit) to support NASA's Mars exploration program set to begin after 2003.
We would like to extend the Earth's Internet to include other planets. After Mars, the next planet we will probably attempt to connect to is Jupiter and its moon Europa. The reason for choosing Europa is that this moon has signs of a liquid ocean underneath the frozen ice cap. One thing that has driven the search for life on Mars and our understanding about life on Earth is that, where there is water, there is life. Where life has evolved, life can be sustained. So from this point of view, Europa could be a prime candidate for future human settlements. We will see the IPN growing outwards from Earth as we explore more and more planets, moons, asteroids, and possibly other stars.

The Earth's moon and a space shuttle might be close enough to run off Earth's Internet, depending on the latency between the Earth and the end destination. For farther out destinations, the IPN project envisions separate Internets, perhaps one for each planet or interplanetary spaceship in transit. By keeping the networks separate, engineers avoid having to make service calls to far away destinations, such as Mars to keep up with technological advances on Earth. Also a dot-com database containing more than 20 million names will not have to be constantly sent to Mars. In essence, the "network of internets" which was discussed previously, will become a reality, with multiple gateways located on orbiting bodies, such as planets, moons, asteroids, satellites, and/or free-flying spacecraft.

The Interplanetary Internet is aiming to be fully implemented between 2020 and 2040. Parts of it can be put into place, piece by piece, as new missions into space are launched. Ultimately, we could see a space borne Internet that could revolutionize how people work in outer space, just as the Internet is changing our more Earth-bound life.

Who knows, one day we may see that space will be commercialized - not only for communication services, but also for mining the asteroids, for the operation of space-based hotels, for manufacturing and medical treatments, and for general tourism.

One thing is certain, and that is that the IPN is a project with no particular end point. There will always be somewhere else to connect, another Star, another Planet, another Galaxy. We cannot actually answer the question "When the IPN is finished, what will it look like?", it is analogous to asking "When the Internet is finished, what will it look like, where can it be used, what can it be used for and by whom?"



Introduction Deirdre Ward
Proposed Implementation Rachel McDonnell
Current Status Micheal Tinney
Problems Emer O'Donovan
The Future Philip Kelly
Editing Rachel McDonnel
HTML Deirdre Ward & Micheal Tinney