I apologize for repeating this posting. Some of it was deleted by mistake, and I didn't want important information to be left out. But you've heard it here, on this website before. You've been told that the pursuit of Altice running fiber to the home is NOT financially plausible. Waiting for the proper permits, digging and repairing the trenches to run the line, lawsuits from AT the inside portion of the device will have an Ethernet jack to connect an existing Wi-Fi router, or one of Starry’s making. (The Starry router, called Starry Station, is already on sale and has a Nest-like feel to it; a touch-screen interface displays circles representing consumption levels of devices in the home and other network information.)
They’ll get service from a transmitter the company is still working on finishing: a two-foot-high rooftop unit able to serve between 600 and 900 customers within a roughly one-kilometer (over 1/2 mile) range.
Beta testers in Boston who seek gigabit wireless service from Starry this summer would need to attach this piece of antenna hardware, a little larger than a soda can, outside of a window.
Pricing and service levels have not yet been announced. But it is expected to be much faster than cable service (90 percent of Boston is served only by Comcast) with upload speeds that match download speeds. It may face additional competition, though: in April, Verizon announced it would invest $300 million to bring fiber-optic FiOS service to the city over six years.
The capital cost of providing the Starry infrastructure is a major differentiator. Whereas building a new fiber or other wired network to reach each housing unit and business in an urban setting can cost $2,500 per potential customer, Starry’s cost might be as little as $25 per customer, says Chet Kanojia, the company’s founder.
The model would not require pulling wires into each apartment or business; rather, it would need just a couple of strands of fiber optic cable to each rooftop transmitter. The company, which intends to be the service provider to customers, would lease fiber from an intermediary such as Level 3 to purchase wholesale bandwidth, just as other ISPs do. “By adopting this strategy we are driving the costs down and adding competition,” Kanojia says.
Installing the infrastructure is so cheap it could allow for a profitable business even if only 5 to 10 percent of potential customers took service, far less than the 50 percent or more that is typically needed to provide enough payback on capital costs, he says.
Starry uses high-frequency radio spectrum—between 37 gigahertz and 40 gigahertz—able to carry vast amounts of data. Some companies already use a simple version of millimeter wave technology to provide wireless Internet access to customers, but existing systems are very limited in that they require one new antenna added to a tower for each new customer. They also require a direct line of sight to work, because such high-frequency signals are very easily blocked by objects—people, foliage, and even rain or snow.
The “active phased array” approach overcomes these obstacles. With these systems, wireless transmissions are distributed over arrays of 16 or more antennas, and dispatched in rapid and very complex pulses that allow one transmitter to serve hundreds of customers at a time. What’s more, these arrays can reach some customers’ antennas that aren’t in the line of sight.
Starry says it has measured speeds from 300 megabits per second to more than one gigabit per second at a range of between one and 1.5 kilometers—even amid rain or snow.
Starry, which has 57 employees in Boston and New York, is prototyping parts of the system in-house. During a visit to his office, Kanojia pulled out a custom-machined piece of aluminum looking a bit like a miniature profile of the Chrysler building. It was studded with holes and imprinted with patterns that will help guide radio waves from the company’s transmitters, he said.
Even though Google Fiber and other companies and municipalities are starting to provide gigabit service, today only about 8 percent of the country has access to such speeds, according to the National Telecommunications and Information Administration. “This could be an alternative that could enable a much more competitive environment, and that would be fantastic,” Reed says.
Industry interest in such technology is snowballing. AT&T and Verizon have announced trials for millimeter wave fixed wireless systems, and major players like Qualcomm, Nokia, Ericsson, Huawei, and Google are working on versions. Samsung is working on mobile technologies using 64 antennas to send and receive signals on ultra-high frequencies. And researchers at New York University have also tested advanced versions for mobile networks. But if a gigabit signal reaches your urban apartment, it might well come first from Starry.