SpaceX Pledge 1Gbps Speed and 25ms Latency from its LEO Satellite Broadband

Rocket company SpaceX has released an update on their plans to launch a global constellation of 4,425 small Low-Earth Orbit (LEO) satellites (a further 7,500 could follow) from 2019, which they claim will support ultrafast broadband speeds of up to 1Gbps and low latency times of just 25ms.

At present most Satellite broadband ISPs tend to operate large geostationary spacecraft, which can handle hundreds of thousands or even millions of customers; depending upon how the limited capacity is managed. For example, Eutelsat’s KA-SAT spacecraft was launched in 2010 (here) and this can handle a total capacity of 90Gbps (Gigabits/sec), although end-users are often offered speeds of up to 22Mbps alongside small usage allowances and latency times are very slow (600ms+ for the whole trip).

Future geostationary Satellites, many of which will start to launch from around 2019, are already talking about boosting consumer download speeds up to “ultrafast” (100Mbps+) territory and offering significantly better usage allowances (example); these would of course remain hindered by high latency times.

However SpaceX, as well as several other companies, are all looking to take on the established Space communication giants by building large constellations of smaller LEO Satellites that would sit at a lower orbit and have a much shorter lifespan. Each Satellite will then beam data directly to gateways or user terminals via the Ka-Band and Ku-Band (they’re also looking at the V-Band).

As a result it’s optimistically predicted that end-user latency times could be as low as 25-35ms (milliseconds), which is better than many fixed line DSL based broadband connections. On the other hand we suspect that the full-trip latency may end up coming in a bit above 25-35ms (factoring latency added from Internet servers, local network equipment etc.).

In this setup end-users would need to install “relatively small” flat panel terminals (the size of a laptop) on their homes, which will use phased array technologies to allow for highly directive, steered antenna beams. Meanwhile the LEOs themselves will communicate with each other using optical inter-satellite links, thus creating a “mesh network” flying overhead that will help to keep the ground connection running continuously.

Patricia Cooper, VP of SpaceX’s Government Affairs, said:

“The SpaceX system will consist of 4,425 satellites operating in 83 orbital planes (at altitudes ranging from 1,110 km to 1,325 km). This system will also require associated ground control facilities, gateway earth stations, and end user earth stations. Using Ka- and Ku-Band spectrum, the initial system is designed to provide a wide range of broadband and communications services for residential, commercial, institutional, governmental, and professional users worldwide. SpaceX has separately filed for authority to operate in the V-Band, where we have proposed an additional constellation of 7,500 satellites operating even closer to Earth. In the future, these satellites would provide additional broadband capacity to the SpaceX system and further reduce latency where populations are heavily concentrated.

To implement the system, SpaceX will utilize the availability of significantly more powerful computing and software capabilities, which will enable SpaceX to allocate broadband resources in real time, placing capacity where it is most needed and directing energy away from areas where it might cause interference to other systems, either in space or on the ground. Because the satellites will beam directly to gateways or user terminals, the infrastructure needed on the ground—particularly in rural or remote areas—is substantially reduced, essentially addressing the “last mile” challenge and helping to close the digital divide. In other words, the common challenges associated with siting, digging trenches, laying fiber, and dealing with property rights are materially alleviated through a space-based broadband network.

SpaceX intends to continually iterate and improve the technology in the system, something that our satellite manufacturing cost profile and in-house launch capability uniquely enables. The ability to modify service as necessary, as well as refresh the technology of the satellite system through iterative spacecraft design changes and phased, continuous deployment, is critical to meet rapidly changing customer demands and responsibly utilize spectrum. This approach will ensure that the system remains adaptable to existing and future customer demands.”

At this point it’s wise to take SpaceX’s talk of delivering “affordable … fiber-like speeds” (corrected to ‘fibre’ in our title because.. English) to each individual user with a pinch of salt. The network being proposed would be technically very challenging to manage, with the realities of network congestion being something that could hamper even the best intentions. The claimed 1Gbps end-user speeds also won’t become viable until full deployment has been reached by the end of 2024.

Apparently each LEO spacecraft would be able to handle an aggregate download capacity of up to 23Gbps (Gigabits per second), which is a lot, especially when multiplied by several thousand. However at the same time this is a network that aims to serve around half of the worlds’ unconnected population and that makes predicting end-user distribution and capacity demand rather tricky.

On top of that Space agencies across the world have already been voicing concern about the growing “space junk problem” from lots of old satellites and related bits and bobs that remain stuck in orbit, which is making it increasingly hazardous for rockets to escape the Earth. Adding lots of smaller LEO Satellites is unlikely to help.

Otherwise the first prototype Satellites should be launched later in 2017 and they will be joined by a few more in 2018, which will enable SpaceX to demonstrate the network and prove whether or not their big claims can stack up to reality.