The Vision of a Lunar Internet. The internet as we know it from Earth offers features that seem to be ideally suited for space communication – including certain amounts of fault tolerance, self-organisation and de-centralisation. However, to our knowledge, no attempt has yet been made to harvest these benefits of the internet in a space environment. In contrast to the internet on Earth, such a space internet is established not for online shopping or emailing, but to transmit telecommand, telemetry and payload data between satellites in space and control centers on Earth. With numerous planned missions heading for the moon, including the European Student Moon Orbiter (ESMO) and its American counterpart called ASMO, discussion is taking place concerning an internet-like space data network for satellites orbiting the moon, called the Lunar Internet (a.k.a. LunaNet for short). The Lunar Internet adopts proven technology from the earth internet – for example its overall architecture, the way data is forwarded, its error correcting algorithms – and adapts it for communication between spacecraft and ground.
The Lunar Internet is reliable. Just like the everyday internet on Earth, the proposed lunar internet has a decentralised approach, with network nodes in every participating satellite and a fair amount of links established not only between a satellite and its ground station but also between the satellites themselves. This shifts the mission-critical importance of certain dedicated communications connections to a far more flexible architecture of multiple possible signal paths. The network can re-organise itself in case of failure of some connections, keeping all participating satellites on line by the best means possible. In this way, it offers inherent redundancies.
The Lunar Internet is cost-efficient. Additionally, multiple signal paths make dedicated and cost-intensive relay satellites, today often needed to ensure continuous spacecraft-to-ground connection, much less important. In the proposed lunar internet (just like in the earth internet), the data packages look for their way to the receiver for themselves. No unobstructed line of sight between a satellite and its ground station has to be present necessarily; only a connection from the satellite to some other on-line spacecraft participating in the network is needed. In this way, the lunar internet also promises to be a very cost-efficient approach.
Complexity can lead to chaotic behaviour. As depicted above, the proposed lunar internet is far more complex than the more traditional links today in use. This is one of its strengths, but on the other hand also holds the potential for unforeseeable behaviour. When a failure occurs while determining possible data routes, the network organisation can quickly become chaotic and the advantages are lost. However, this risk can be minimized through intensive testing and simulation during the design phase, taking into account experience from the earth internet. To minimize the impacts of unforeseen errors that remain in the Lunar Internet, a traditional, dedicated communication connection can be used for direct data transfer between ground station and satellite (not using other satellites) as well.
A Space Internet can increase latency. In comparison to direct data links, a Luna Internet requires some additional information in its data packet headers, so a bit more data is to be transmitted and latency is slightly increased. As most data is transmitted in packets of major size (compared to the header size) and most operations are not time-critical to the highest degree, this disadvantage will be negligible in most cases.
A Space Internet can minimize latency. Under different circumstances, however, latency can also be decreased compared to today’s communication links. For sophisticated operations needing a data link between two satellites, the data to be transmitted does not have to be routed down to earth and up again to the destination satellite, but can directly be sent from one satellite to the other, minimizing latency through optimized routing.
Possible extension of a Space Internet to the Ground Segment. The possibilities described above mainly relate to the space segment. However, the concept can be extended to the ground segment as well, thus generating additional benefits. In today’s common practice, usually at least one dedicated ground station per satellite is needed, at least during times when data has to be transmitted or the satellite has to be observed especially closely. This increases the mission’s overall effort and cost. The Lunar Internet, or a space internet in general, can ease this burden considerably. Multiple missions can share a relatively low amount of ground stations and use them jointly. In this concept, each ground station establishes a link to one of the satellites, and the link to the satellites not having direct contact to ground can be established by satellite-to-satellite data routing. This concept does not mean considerable loss of reliability or safety, because the lunar internet offers inherent redundancies and there still is the possibility to use direct links during critical mission phases as well.
SSIMUC and LunaNet. LunaNet will be a new protocol specification, allowing satellites to exchange data in an internet-like way. ESMO and ASMO are inteded to serve as a test platform for this promising technology. Our task is to implement LunaNet in order to test its basic functionality and demonstrate its features. If ESMO and ASMO behave well and fulfill the expectations, they ultimately can serve as foundation for a fully operational Lunar Internet, offering versatile communication services to future missions to the moon.
Implementation. Thanks to support by experts of the German Aerospace Center's Institute of Communications and Navigation, a protocol simulator was set up. It can simulate the behaviour of a communication channel such as bit errors and delays. This simulator is intended to serve as framework to develop, implement and test the LunaNet communication protocols. A screenshot can be found here.