DARPA, lasers and the internet in orbit

Statellites are critical military infrastructure for espionage and communications. They are also vulnerable to attacks and outages. In November 2021, three months before invading Ukraine, Russia fired a missile at a defunct satellite. Then, in October, a Russian diplomat said commercial satellites could also be legitimate targets. The satellite systems used by Ukraine have been hacked and blocked. Ground antennas have been attacked.

In light of this sort of thing, the US military establishment is concerned that its satellite network is not up to par. But he has a plan. The space-based adaptive communications node (Space-BACNor Space Bacon, for his friends) will, if successful, create a laser-enabled military Internet orbiting the Earth piggybacking a number of satellites that would have been launched anyway.

Space Bacon was born from an idea by DARPAspecial projects research arm of the Department of Defense, and is an intriguing orbiting echo of the original, Earth ARPANET, which evolved into the Internet. (He was so called at one point DARPAs history when the organization lacked the initial d for defense.) The plan is to equip as many newly launched satellites as possible with laser transceivers capable of communicating with counterparts up to 5,000 km away. Satellite owners will pay for these transceivers, but then receive payments from the US government for their use.

Zipping fantastic light

Space Bacon promises many benefits. Unlike radio, the normal way of communicating with and between satellites, laser beam transmissions are difficult to intercept and nearly impossible to jam. Indeed, adversaries may not even know when a transmission is in progress, an advantage for operational secrecy.

Lasers also offer much higher data rates than radio waves. Some satellite constellations already use lasers for communication between members, and these reach about two gigabits per second (about 200 times what radio can handle). DARPA, however, has asked Space Bacons contractors to develop equipment capable of transmitting, in a single beam, 100 gigabits per second. This is enough to send several high definition movies at that time.

The ability to transmit military information from one bird to another in this way and without the constraints imposed by differences in the communication protocols of the satellites providing piggyback will greatly simplify matters. Individual satellites can only download data when they are within range of a ground antenna belonging to their particular network, or via another member of that network, which is likely to be in a similar orbit. A satellite in the Space Bacon system, by contrast, can transfer data to another, possibly belonging to a different operator, in a different orbit. And that satellite could, in turn, hand it over to yet another, until a suitable ground antenna is on hand.

At present, delays in reporting caused by lack of network interoperability mean, for example, that a tank detected by a satellite could have departed before its position has been received by anyone who can use the information. Space Bacon will more or less eliminate this latency.

It will also offer one of the vaunted benefits of the original ARPANET design, which the Internet has inherited. This is the automatic redirection of a message if a node (i.e. a particular satellite or ground station) is disabled. Furthermore, by bringing almost any relevant ground station into play, particularly sensitive data can, as Space Bacons program manager Greg Kuperman observes, be routed through antennas to locations where attempts to intercept the latest radio transmission, trip are considered less likely.

At the heart of this is the aiming accuracy that will be incorporated into the Space Bacons lasers. Phil Root, head of DARPAs Strategic Technology Office, says thinking about this drives me crazy. Satellites in low Earth orbit (LEOs, those below an altitude of 2,000 km, and the kind Space Bacon will use to start with) travel at about 7.8 km per second, often rolling as they go. Connecting the optical heads on two of these will be an epic task. The progress, however, has been impressive. Mynaric, a company based near Munich that is designing heads for Space Bacon, can adjust the trajectory of a laser by just 57.2 millionths of a degree. At a distance of 1,000 km, this results in a beam shift of less than one metre.

To sweep its lasers smoothly, Mynaric uses a system of motorized lenses and tilting mirrors. Another contractor, mBryonics of Galway, Ireland, uses electronic signals to alter the phase of lights by fine-tuning the direction of the beams in a way analogous to redirecting photons through a glass lens.

However, even with these levels of accuracy, direct first hits to a distant head are likely to be rare, says David Mackey, mBryonics’ chief technologist. Optical heads attempting to connect will then conduct what he calls blind spiral search patterns. When a beam finds its mark, the signal will inform the receiver of the senders exact location. Using a different wavelength, to avoid interference, the receiver will then send a laser along the same path to confirm the connection. Mr. Mackey thinks his kit will complete these orbital handshakes in ten seconds.

Current laser-based satellite communications rely on sensors called photodiodes. Space Bacon’s higher data rates require a different approach. The message-carrying photons will enter a single strand of optical fiber with an aperture only ten microns in diameter, much smaller than a photodiode’s 100 microns, of which more than one may be needed anyway. Mybryonics hopes to accomplish this trick by using a mirror with a complex curved surface to redirect photons into an iPhone-sized device that focuses light and shoots it into the fiber optic filament.

The Space Bacon Systems specification requires that you draw no more than 100w from its host satellite. This limit poses a problem to the processor needed to translate between the different data protocols used on satellites, for only 40w of that 100 is available to do so.

One company working on this is Intel, an American chip maker. He is designing what Sergey Shumarayev, the project leader, describes as a Rosetta Stone-type modem. Mr Shumarayev says commercially available components were left out to do this because they are too large and power hungry. He calculates that if they were used, the result would be as big as a pizza and they would consume 400w. His team is trying to shrink the size of the pizza in a matchbox by using what they call chiplets, instead of larger semiconductors.

Bringing home the BACN

DARPA he wants Space Bacon to cost a maximum of $100,000 a satellite, the best way to encourage participation. It bodes well that Amazon, SpaceX, and Viasat, a lesser-known but well-established satellite communications company, are all designing command-and-control architectures for Space Bacon.

Amazon plans to launch a broadband LEO constellation of satellites called Project Kuiper. This can incorporate Space Bacon transceivers. SpaceX could add them as it expands Starlink, its broadband LEO constellation, from its current complement of some 3,500 satellites to tens of thousands. The existing network of Viasats is different. It is based on five large satellites in geostationary orbits 36,000 km above the Earth, with three more to be added over the next 14 months. The company also has an extensive network of ground stations to bring to the party.

A command and control problem is the elaboration of optimal paths for data transmission. Craig Miller, head of government systems at Viasat, says it’s more complicated than solving the traveling salesman problem, a math classic. As the network grows, calculating the most efficient routes becomes increasingly difficult, and not just because the nodes will move. Mr. Miller’s team must also assign and computationally account for varying confidence scores for potential data jumps, as some are more vulnerable than others to enemy action.

DARPA it plans to select the best subsystems this summer and hopes to have a prototype ready for testing LEO before 2025. If all goes well, the network could be extended to geostationary orbits. Allies, Dr. Root reckons, could be invited to join. The Americas’ opponents will no doubt be watching closely.

Curious about the world? To enjoy our mind-expanding science coverage, subscribe to Simply Science, our weekly subscriber-only newsletter.