I recently ran into Paul Callahan at Interop New York. I've known Paul for a long time; he was one of the founders of AutoCell Labs, who developed a technology for automatically configuring and optimizing Wi-Fi that I consider to be sheer genius. It really works, but it so far has not caught on; something about added cost, non-standard elements, and other mostly-valid but completely short-sighted considerations. Anyway, Paul is now VP of Business Development at Airvana, a company best known for EV-DO base stations but which really has a much broader view of the wireless world. Paul is a huge fan of femtocells, and said some fairly negative things about public-access Wi-Fi. I still think the latter is going to be a major success, and Paul didn't change my mind. But he did invite me to visit Airvana to see what they're doing in femtocells, and I was quick to take him up on the offer.
So, I recently met with a number of folks at Airvana, most notably Sanjeev Verma, who is Founder and VP of Marketing and Business Development. Airvana has great technology, products, and especially people - and Sanjeev made some excellent arguments about femtocells, undeniable in their truth. First, carriers like femtocells because they appeal to their installed based and solve the problem of wireless reliability within the residence. They also further lock in the consumer via equipment purchase and generate additional monthly fees, and the added cost makes it unlikely that a single household will deploy multiple femtocells supporting multiple carriers. He sees the predominant technologies in femtocells as being CDMA and UMTS, not WiMAX, and noted that Sprint is already deploying them under the AIRAVE brand in parts of Denver and Indianapolis, offering unlimited calling from home as part of the monthly fee. Femtocells could obviate the need for cellular/Wi-Fi convergence (like T-Mobile's HotSpot@Home), and allow the use of existing single-radio handsets with not modifications to the existing cellular footprint.
I can't argue with any of this, other than the part about cellular/Wi-Fi convergence, which I still think is a better approach. But I got to thinking that there is a potential flaw in the femtocell vision, and that stems from the lack of network neutrality here in the US. I think net neutrality is essential to the future growth of broadband - not the growth of the carriers themselves, who can do just fine regardless, but rather the services available via their networks. The carriers will continue to push the idea that they get to define what information moves over their networks, and whom their customers can connect to. This is the height of arrogance, IMHO, as the carriers are using public airwaves (we still own them; they just license them) and public rights-of-way for cabling. They should therefore be required to carry any legal traffic. If they want to charge for priority service, fine, but they can't refuse a connection outright. And, yet, today, they can. And, yes, this could also threaten converged services in the residence as well.
So, which brings me back to my original point - suppose a broadband carrier, who, after all, wants to sell their own phone service, decides that femtocells are competitive, and they simply refuse at some point to forward, or otherwise interfere with, traffic to and from femtocells. Just suppose.
Mathias is a principal at Farpoint Group, a wireless advisory firm in Ashland, Mass.
|
|
The Flaw in femtocells
I cannot speak for any other company's products, but Samsung's Femtocell: Ubicell communicates to the WSS via IPSEC so will be impossible for the broadband carrier to differentiate between femto traffic vs. corporate communications traffic. That said, the codec for CDMA is very efficient and does not use much bandwidth, (8K or 13K) so the femtocell user does not require nearly as much traffic as someone running a voip or video session.
If you have any questions or need add'l info. let me know.
Regards,
Jim Parker
Re: The Flaw in Femtocells
It's not a question of the amount of traffic, but rather the nature of the traffic. A DSL/cable/etc. carrier could recognize competitive traffic (if not encrypted) by deep-packet processing and certainly by the destination address.
Thx. Craig.
Licenced Spectrum vs Open Access
The key point here is that femtocells run on licenced spectrum, rather than the unlicenced free-for-all that is WiFi. Operators rightly expect to make profit and take some financial benefit from the large fees they have paid goverments for exclusive use of their frequencies.
The benefit to the end user of controlled/licenced spectrum is that this is properly co-ordinated, planned and administered by mobile/wireless operators to ensure maximum capacity/coverage and thus best service for their users.
Femtocells fall into a category where they are more likely to be semi-autonomous, sensing what frequencies should work best in any given location to minimise interference with the external macrocells. I'd expect this to give better service than WiFi in congested areas.
Many femtocell vendors are building complete network subsystems into their products which could allow them to completely handle and offload the (data) traffic directly to the internet - effectively enabling a local WiFi hub. What will be interesting to watch develop is whether operators will permit users unlimited data tariffs when using their systems in the home (which effectively have no incremental data traffic costs), and thus enable replacement of WIFi service which works inside and outside the home.
Mostly likely, this would be limited to subscribers of that network operator, but it doesn't have to be.
Read more on the potential business cases, opinion s and technical information about femtocells at www.thinkfemtocell.com
Re: Licenced Spectrum vs Open Access
It's not a question of which wireless spectrum is being used, but rather whether a broadband carrier might refuse to carry the traffic because it's competitive with a service they sell. This also has nothing to do with incremental traffic or even the amount of traffic - it's purely a business decision.
And I don't think we want to replace Wi-Fi, but that's a topic for another day...
Thx. Craig.