This past week, I was fortunate enough to attend the Grace Hopper Celebration for Women in Computing, a national conference with 18,000 attendees that revolves around promoting women in tech. While I was there, I attended a workshop hosted by X, the moonshot subsidiary of Alphabet, to learn about the inner workings of Project Loon. During this workshop, Puerto Rico received great news: Project Loon has been approved to provide emergency cellular service to the island in the aftermath of Hurricane Maria. Here’s a summary of what I learned:
The Magic Behind the Balloons
50% of the world’s population still does not have Internet connection. The reason? Cell towers are simply not designed for rural/remote areas. It’s expensive to drag power and fiber across the globe to these areas. On the contrary, in developed areas like the U.S., the typical cell tower can transmit coverage to a surrounding area with a diameter of around 3 km. This coverage area simply depends on the height of the tower: as the tower increases in height, it can cover a wider area. Because of this, X places balloons 20 km high in the sky, which allows them to cover a ground area with a 80 km diameter– a vast improvement from a typical cell tower.
The main premise of the project is the following: the balloons harness solar power while flying in the stratosphere. The balloons are directed to their desired position by “riding the wind.” By predicting wind speeds and directions, X creates algorithms and decision-models to direct balloons through wind currents to get them where they need to be. Once the balloons are in their appropriate position, X partners with local telecommunications companies to provide the LTE connection. Sometimes acquiring these partnerships is difficult if the area is extremely remote, such as in a rainforest. When the balloons are in place, they are designed to stay afloat for 100-200 days in order to provide months of service, but not too long such that the communication protocols will become outdated.
So how exactly does Internet get from a floating object to a farmer accessing their email in rural India?
- Let’s pretend that I’m sending an email to this farmer. When I click send, a wireless Internet signal is sent to a balloon in the vicinity, which picks up the email via LTE. The email then hops across the balloon network from one balloon to another. If it needs to travel across the ocean, it’ll be routed through a nearby ground cell tower and sent through fiber cables that are laid underneath the ocean. Otherwise, the message will be beamed down from the nearest balloon to the person in the remote area. This all sounds easy enough, but what’s the challenge?
While the process itself sounds simple, enabling balloons to speak to each other isn’t exactly a walk in the park. Balloons are strategically positioned far enough away from each other to maximize coverage, but this makes it difficult to pass messages between them. To illustrate this, picture the following analogy: Imagine that you’re standing in the end-zone of a football field with a laser pointer in your hand. You need to aim at a grain of rice that is in the other end-zone. Seems difficult, right? Unfortunately, it’s even more difficult than that: now imagine that you are floating in the air with your laser pointer and the grain of rice is also floating, but you both might be jostled by the wind. If your aim is good enough to hit this grain of rice, you should probably reconsider a career in tech and take up archery or something more athletic. The receptors on the balloon are not easy to hit and researchers have developed breakthrough technology through AI to solve this challenge. Luckily, while controlling a floating piece of plastic to send these messages is a challenge, after years of testing they’ve finally reached a breakthrough.
The Resurrection of Peru
In May, X launched balloons in Peru after floods devastated homes and roads in the Inca Region. X had already been partnering with telecommunications company Telefónica for months to test their balloon capabilities. When disaster struck, the infrastructure was already in place for them to begin delivering Internet connection at a mass scale. The balloons provided 160GB of data to tens of thousands of users, which is the equivalent of 30 million WhatsApp messages, or two million emails. The coverage area also spanned 40,000 square miles, which is the equivalent of the size of Switzerland. This was the first time Loon had successfully generated connectivity for more than small-scale testing purposes, and proved the viability of the project.
This event illustrated the impressive capabilities of the project. Balloons lasted for 98 days in Peru, while drones created by Facebook’s Aquila drone have only lasted in flight for 96 minutes. Peru shed light on a new opportunity for disaster relief, and now Hurricane Maria has provided them with a similar situation.
Saving Puerto Rico
As of Sunday evening, 82% of Puerto Rico is still without cell service. In a time like this where millions of Puerto Ricans are still without needed communication services, all ideas for enabling connectivity are needed. On Friday, the FCC granted Project Loon a temporary license to distribute connectivity to the Caribbean island. According to FCC Chairman Ajit Pai, “That’s why we need to take innovative approaches to help restore connectivity on the island. Project Loon is one such approach. It could help provide the people of Puerto Rico with access to cellular service to connect with loved ones and access life-saving information.”
The current challenge is finding a telecommunications company to work with them on the ground. Spokesperson Leahy Libby of X said that “To deliver signal to people’s devices, Loon needs be integrated with a telco partner’s network — the balloons can’t do it alone. We’ve been making solid progress on this next step and would like to thank everyone who’s been lending a hand.” The difference for X this time is that they have no existing infrastructure in Puerto Rico like they had in Peru. They’re essentially starting at square one as they attempt to acquire a telco company to work with. While they may have tackled the difficult technical challenges of enabling LTE through plastic, the project still faces the basic hurdle of finding a partner in this devastated region. The lesson? Perhaps Loon should focus more on building partnerships to be prepared for future situations in disaster-prone areas.
While the project may initially have been developed to increase the breadth of Internet connection, there may be a new niche market. This disaster recovery model can possibly be expanded into helping deliver status updates by maintaining digital geographic grids or sending alert messages to loved ones. It’ll be interesting to see how Loon progresses in the upcoming years, especially as climate change affects the prevalence of natural disasters.