ITEM: Facebook has officially given up on its project to deploy broadband UAVs that would connect the unconnected to the internet from the air. Or at least it’s giving up on doing the UAV part.
Four years ago, Facebook launched a program called Aquila with the objective of developing a high altitude platform station (HAPS) system that could provide 24/7 broadband connectivity from the belly of a UAV hovering above areas with little or no terrestrial connectivity options for months at a time. Essentally Facebook started the project from scratch, designing everything from the HAPS communications payload, gateways, fiber PoPs and access points to the UAV itself.
The basic idea was to design a network using UAVs to do what satellites do – provide transponders to relay comms signals to and from the earth – only far more cheaply and closer to the ground. Relaying internet traffic via satellite creates latency issues because of the physical distances involved – drones have negligible latency issues, and – unlike satellites – you can bring them back in for refuelling and maintenance whenever you like.
However, four years later, Facebook’s Aquila project hasn’t made much progress – the team have managed only two test flights since 2014, the first of which resulted in a crash landing. Turns out that the UAV is the hard part – something that Google also found out the hard way. It bought UAV manufacturer Titan Aerospace in 2014 with a similar plan to turn UAVs into broadband base stations. Google shut down Titan at the start of this year.
To be clear, Facebook isn’t giving up on the idea of broadband drones entirely – it’s just giving up on doing the drone part in-house. According to a blog post from Facebook’s director of engineering Yael Maguire, the official reason for dropping UAV development is that proper aerospace companies are now investing in new UAV technology suitable for broadband comms – something that very few aerospace players were seriously considering in 2014.
Put simply, now that actual aerospace experts are on the case, there’s no need for Facebook to blow money trying to do it themselves.
Also, Maguire adds that while the UAV part has proven tricky, the comms part (which uses mmWave technology for air-to-ground and point-to-point communications) has made considerable progress – to include achieving symmetrical data throughput speeds of 40 Gbps at a distance of 7 km. Consequently, Facebook intends to continue to develop its HAPS technology with partners, he said:
Going forward, we’ll continue to work with partners like Airbus on HAPS connectivity generally, and on the other technologies needed to make this system work, like flight control computers and high-density batteries. On the policy front, we’ll be working on a proposal for 2019 World Radio Conference to get more spectrum for HAPS, and we’ll be actively participating in a number of aviation advisory boards and rule-making committees in the US and internationally.
That makes sense, given that Facebook’s track record in network communications tech development is surprisingly good. Its work with the Telecom Infra Project has produced a number of interesting and (potentially) viable open-source networking products, from its TerraGraph fixed-wireless access technology and Voyager DWDM white box to micro LTE base stations, fronthaul, network slicing, optical network planning and (obviously) artificial intelligence. The expertise that Facebook is developing within TIP (which means, by the way, that it is developing telecoms tech in collaboration with telecoms networking experts) can also serve its Aquila HAPS tech.
All they need is for someone else to figure out how to actually make the drone part viable.