Launching the Internet of Things

Last week, on December 6th, the tenth flight of RocketLab’s Electron lifted off from the pad in Mahia, New Zealand. Irreverently nicknamed as “Running Out of Fingers,” the flight was a major milestone for the smallsat launcher company testing procedures that might allow RocketLab’s 55-foot-tall Electron rocket to become the second reusable booster on the planet.

The rocket lofted seven spacecraft into low Earth orbit, six of the seven satellites onboard were part of a cluster of PocketQube satellites from the Glasgow-based satellite startup AlbaOrbital, called Alba Cluster 2, and amongst them was FossaSat-1.

Built by Julian Fernandez, Richard Bamford, and Chris Mulhall, the FossaSat-1 satellite takes advantage of both the plummeting cost and integration of off the shelf chips and radios that is the “peace dividend” of the smartphone war.

Smaller even than the now ubiquitous CubeSat standard, FossaSat-1 is what is known as a PocketQube  —  measuring just 5 × 5 × 5cm and weighing in at only 250g, it will be one of the smallest satellites ever launched into space.

The satellite is intended to act as a LoRa station in Low Earth Orbit. Designed for low-power, low-bandwidth, and long-range applications—10s of km on the surface, with far longer distances possible assuming line of sight — the LoRaWAN standard is intended for machine-to-machine communication, and the Internet of Things and is especially useful for large scale distributed sensor networks. It’s also the current front runner to win the battle in the low-powered Internet of Things wireless standards wars. In part due to the cheap and ready availability of hardware, and open source software.

Placed into a 400km Sun-synchronous Orbit (SSO) the satellite uses the UHF Amateur Satellite band (435–438 MHz) and carries both a LoRa beacon and repeater onboard. Since it is in a polar orbit and is effectively providing ‘ball of yarn’ coverage of the whole planet — albeit not at the same time — the satellite is perfectly placed to relay sensor data from planet wide sensor networks back to Earth-based LoRa networks.

While this isn’t the first time LoRa signals have been received from space, it is the first time  —  at least as far as I know  —  where a satellite dedicated to dedicated LoRa transmission has been launched, and unlike many satellites you don’t need to make a big investment to build or rent a ground station. Instead the LoRa signals can be received by ground station hardware that can be built, or bought, straight off the shelf for under a hundred dollars.

In fact, if you’re intending to base you ground station around the TTGO-LoRa32 V2.0 board, you should be able to download Ground Station code straight from GitHub and everything should work out of the box after doing a little bit of configuration.

Unfortunately, despite a successful launch there were problems during deployment of FossaSat-1. According to Fossa Systems, “FOSSASAT-1 is shown to be working correctly and healthy. However, due to the antennas not having deployed, the signal is very weak and can only be picked up with large antennas. LoRa remains not functional for small stations due to this…” however they are still working on deploying the antennas.

This means that while cheap off the shelf hardware can’t (yet) receive signals from the satellite, LoRa signals are being received by ground stations with higher gain antennas.

However, even if Fossa Systems doesn’t manage to sort out the problems with FossaSat-1, it is also intending to launch two more LoRa satellites in March next year to expand the LoRa network coverage on orbit.

The launch of FossaSat-1 highlights what I personally think is the advantage that LoRa has over its competitors in the low-power, low-bandwidth, protocol wars. The ability for anyone to build their own Internet of Things LoRaWAN network, and integrate it into the community without replying on third-parties like network providers. Albeit in this case it did rely on one third-party provider, that’d be the launch company.

It’s also really rather starting to look like Jeff Bezos’ statement that “…there’s not that much interesting about cubesats” will turn out to be the twenty first century’s version of “…nobody needs more than 640kb.”

Because the idea that these cheap and cheerful satellites—it costs little more than a family car to put a PocketQube satellite into orbit—are going to make a big difference to the Internet of Things is now pretty clear.

After all, FossaSat-1 — and beyond it the potential for a small constellation of LoRa satellites like it — is really just a low-cost, open-source, open hardware version of the larger smallsat constellations that are coming online now, like SpaceX’s Starlink.