If Amazon is Testing Their Own Wireless Network, $10 Says Its Not for a Consumer Product
Bloomberg is reporting this evening that Amazon wants to become an MVNO. If this rumors is true then Amazon’s next tablets could browse the web via Amazon’s own satellite network.
The wireless network, which was tested in Cupertino, California, used spectrum controlled by satellite communications company Globalstar, said the people who asked not to be identified because the test was private.
The trial underlines how Amazon, the world’s largest e-commerce company, is moving beyond being a Web destination and hardware maker and digging deeper into the underlying technology for how people connect to the Internet. That would let Amazon create a more comprehensive user experience, encompassing how consumers get online, what device they use to connect to the Web and what they do on the Internet.
It’s a bit of a surprise that Amazon wants to get into this industry, and that goes double when you consider who Amazon is working with. Globalstar isn’t a cell company; they own a network of satellites and lease space to consumers as well as corporate customers (even though they have been pursuing regulatory approval for mobile broadband).
If someone had told me yesterday that Amazon was going to build a wireless network I’d have thought they would lease bandwidth from AT&T, Verizon, or someone. That is a fairly common practice in the industry with a dozen or more of the smaller players buying from the handful of major networks.
But Amazon is instead going for a satellite network instead of a cell network. Could this be a sign that Amazon plan to incorporate a satellite data plan in their next tablet?
I doubt it.
The latest info on 3G mobile data use shows that tablets make up only a small fraction of activity. The vast majority of bandwidth (70% plus) is being consumed by the iPhone and by Samsung’s smartphones.
Tablets tend to get more of their data via Wifi, so if today’s rumor has to do with a consumer product then it is most likely going to be a smartphone. But I don’t think that’s it either.
If Amazon was going to do a smartphone then they would probably lease bandwidth from a cell network like AT&T, Sprint, or Verizon. It’s cheaper and more reliable.
If nothing else I doubt they plan to base a product on Globalstar due to the fact that it can’t guarantee service (according to Forbes):
Globalstar went bankrupt in 2002. It has lost about half of its first generation satellites and is just trying to rebuild itself. Presently, its service is very limited and not entirely reliable on a 24/7 basis.
Or is that quote wrong? A reader has pointed out that Globalstar’s second-gen satellite network is already fully deployed and that it has nearly the same coverage as the first-gen network.
On the other hand, Globalstar is trying to get regulatory approval to get into the (ground-based) mobile broadband market, which is what Amazon would need in order to support a consumer product. Here’s the WSJ from last year:
Globalstar Inc., a small satellite communications company, has filed to use its airwave licenses to provide mobile broadband services, a move that could eventually make the wireless spectrum more valuable.
The Covington, La., company filed a two-prong plan with the Federal Communications Commission to use the spectrum, something that would require a partnership with a larger wireless data provider or telecom company. As smartphones and data consumption have exploded, expanded access to airwaves has become important to wireless carriers and has become a stated priority of regulators.
Globalstar has tested the possibility of using their frequencies for mobile broadband but doesn’t yet have that regulatory approval, and it’s not completely clear that they’ll get it.
Of course, that WSJ story brings up another interesting quirk in this rumor. That company has a market capitalization of under $100 million. Amazon wouldn’t need to lease bandwidth; they can afford to buy Globalstar outright.
In any case, if Amazon is looking to use Globalstar for a terrestrial network then that network would have to be built first. That could likely delay the launch of the consumer products on the network by at least a year if not longer.
And that’s why I tend to think it’s not a consumer product. What do you think?
image by wharman
The Commons August 22, 2013 um 11:03 pm
>It has lost about half of its first generation satellites and is just trying to rebuild itself. Presently, its service is very limited and not entirely reliable on a 24/7 basis.
This is irrelevant because Globalstar is asking the FCC to let it use the spectrum it controls for ground-based purposes.
Nate Hoffelder August 22, 2013 um 11:20 pm
Yes, but do they have that regulatory approval yet?
Edit: not yet.
The Commons August 23, 2013 um 12:02 am
No, because one thing it’d have to demonstrate is that ground-based operations wouldn’t interfere with other services. The comment was merely to nip any speculation satellite service ("But Amazon is instead going for a satellite network instead of a cell network.") in the bud. No need for any doubt; satellite data service on the CE device-level is slow and high-latency. Concerning running a wireless network, Globalstar’s spectrum is probably worth more to Amazon than whatever hardware it has floating up in low Earth orbit.
Side note: that Forbes article is 5 months old and even then neglects to mention that its second-generation satellite set was fully operational… and had been for a month. No idea where that information was from, but the post links to a seven-month-old YouTube interview that he did. So, connect the dots. All in all, that post is quite irrelevant.
An interesting thought about that hardware, though: Bezos IS the founder of Blue Origin, a private aerospace company…
Nate Hoffelder August 23, 2013 um 12:13 am
I did see on Wikipedia that they had a second gen satellite network, but it also said that the network was deemed complete after only 24 of the 48 had been launched.
The Commons August 23, 2013 um 12:31 am
Their coverage map seems comparable to Inmarsat (which manages with less than a dozen) with the main exceptions of southern Africa and India, and more importantly their Gen1 and Gen2 coverage maps look roughly similar: i.imgur.com/UPeYdou.jpg
Maybe they just figured out that they don’t need 48 satellites to provide the same coverage?