Small cells

Apple Goes Patently Femto

6:40 AM -- Apple Inc. (Nasdaq: AAPL) has filed a patent at the U.S. Patent and Trademark Office for femtocell technology, which was spotted by Macnews.

No doubt this discovery will be good news for small-cell folks, happy to have Apple and femtocell together in the same sentence. But beyond that, why does Apple want a femto patent?

The invention is called "wireless femtocell setup methods and apparatus," which appears to be a technique for reducing interference. According to the abstract, it's described as "Methods and apparatus that enable a wireless femtocell to operate in its designated frequency so as to minimize interference between the wireless femtocell and neighboring base stations (and other femtocells or nomadic cells)." Get the full details here.

Such a technology could be applied not only to public-access small-cell deployments to reduce interference with macro base stations, but also to an indoor scenario to prevent a femto device from messing with a Wi-Fi access point.

The existence of the patent doesn't necessarily mean Apple is doing something with small cells. But what if Apple does indeed have designs for the tiny base station technology? If so, where would it make sense for Apple to use small cells?

Aditya Kaul, practice director for mobile networks at ABI Research , wrote in a recent blog that Apple already sells devices to which it could add femto support -- such as the AirPort Express Wi-Fi base station.

But the bigger question to speculate about is: With small-cell technology integrated into its products or software, would Apple be able to go around mobile operators and provide wireless connectivity services directly to consumers in their homes?

Also, given how litigious the mobile sector is now, Apple's patent move is bound to make small-cell pioneers like ip.access Ltd. and Ubiquisys Ltd. nervous.

-- Michelle Donegan, European Editor, Light Reading Mobile

Kit Kilgour 12/5/2012 | 5:37:40 PM
re: Apple Goes Patently Femto

The main claims of this patent seem somewhat questionable, in that the techniques claimed were already deployed in 3G trial networks in the USA at the time the patent was filed, as well as being in use in 2G picocell deployments in the USA and worldwide.

Within 3GPP standardisation discussion, the techniques, known as network listen (NWL) or similar were already discussed as being clearly needed for a femtocell to set itself up, and there are probably FemtoForum (now Small Cell Forum) documents from before. the date. As Apple did not attend the 3GPP femtocell standardisation discussions, nor is a Small Cell Forum member, it is possible that the engineers filing were just not aware (and clearly neither were the USPTO). All the major femtocell vendors will be unaffected by this patent as they would be able to demonstrate prior internal use. The only area not considered is the business process aspects, which are not patentable in many countries.

I would be much more interested if Apple turns out to have filed a significant number of small cell-related patents.

Kit Kilgour

(Head of Standards, ip.access Ltd)

The opinions expressed are my own and do not necessarily represent those of my employer


gpricejones 12/5/2012 | 5:37:37 PM
re: Apple Goes Patently Femto


I also totally agree on the prior art and use of such frequency scanning mechanisms, I had numerous discussions back in early 2007 with most of the suppliers in this space, all of whom were promoting such capabilities within their products.

Additionally the patent continuously refers to frequency allocation, where as in (the referenced) UMTS/3G systems actually the whole network operates at the same frequency (or multiple bands) of two paired 5Mhz carriers, then uses scramble codes to allocate to specific cells (including femto cells), unlike GSM 2G, which certainly did use frequncy planning and allocation of 200Mhz carriers.

So all in all a pretty amateurish attempt at claiming something that they really didn't invent. Hope that their iPad patents are more water tight.


<div>opinions of this individual.</div>



Michelle Donegan 12/5/2012 | 5:37:33 PM
re: Apple Goes Patently Femto

Yes, it does seem strange. It was first applied for in 2008 and just granted now. Is that a normal timeframe for patent approval?

Also, I'd like to know who does hold the patents for this kind of technology, then.

Michelle Donegan 12/5/2012 | 5:37:33 PM
re: Apple Goes Patently Femto

Thanks, Kit. That's interesting and makes it sound like this patent is meaningless. Is going to court the only way to test its merit -- innovation by litigation? Although I can't see why that would be in Apple's interest for this kind of technology.


Kit Kilgour 12/5/2012 | 5:37:29 PM
re: Apple Goes Patently Femto

2008 to 2012 does seem to be an exceptionally long time to grant - perhaps a lot of discussion between inventors and the examiner trying to argue novelty?

As the filing was in the US there is the distinction between date of invention vs. date of filing, so there may have been some delay. &nbsp;As we completed the 3GPP Rel-8 specifications in c. February 2009, and there are public drafts in late 2008 there can't have been much delay if the examiner was being thorough in searching 3GPP tdocs.

I don't know US procedure, but in the UK, patent applications (whether granted or not) are published c. 18months after submission unless they are withdrawn.

Kit Kilgour 12/5/2012 | 5:37:29 PM
re: Apple Goes Patently Femto

The only place to go to court would be the USA (I assume).&nbsp;

As regards patentability, there is the question as to what was either 'public' or 'widely known in the industry' at the time of filing. Network Listen aka 'sniff mode' or similar &nbsp;has been known since at least 2001 (There are some Ericsson GSM-on-the-net' patent groups), and ip.access was testing 2G solutions with it built in. Clearly if a conference presentation or granted patent exists before May 2008 then it is demonstrably meaningless, and it is a matter of the USPTO examiner not being aware/finding the relevant documents.

I do not know enough of US patent law to know how any request for re-examination would work (or cost). &nbsp;The fact that nobody has spotted a European patent grant to Apple based on the US priority date implies that, given the proscribed time to do this has now passed, Apple either didn't bother to extend worldwide, or they tried and European patent examiners did not grant due to sufficient prior art that they found. &nbsp;

I have had a brief check and there were FemtoForum (now Small Cell Forum) minutes in December 2007 that have discussions on the need to radio 'sniff mode' to enable autoconifiguration for OA&amp;M. These documents are not in the public domain, but any vendor member of the Forum by May 2008 might be able to argue that it was already planned into the product, and of course internal design documents.&nbsp;

I would doubt that anyone is going to bother unless Apple try and enforce it. If they tried to register it as Essential IPR with 3GPP then there may be push back. Otherwise I would have thought that the dominant femto vendors - ip.access, ALU, Cisco and Ubiquisys would just get out their 2007/8 or earlier design documents if Apple approached requesting licensing.&nbsp;


Michelle Donegan 12/5/2012 | 5:37:28 PM
re: Apple Goes Patently Femto

Thanks for sharing your insights here. Sounds very much like Apple wouldn't have a case if it tried to enforce the patent.

The other side to this story is the implication for Apple strategy and product development -- what is Apple doing with femtos? Would love to know if they're actually working on something. I'll work on that...

Kit Kilgour 12/5/2012 | 5:37:26 PM
re: Apple Goes Patently Femto

One further comment is that the initial claim of this patent seems to imply a scenario in which the femtocell asks the 'core' of a network for permission to use a frequency and other parameters, &nbsp;whilst in the standards, the femtocell asks a management system (TR-069 ACS - Configuration Server). This may me an important distinction for any company that was not already implementing a standards-type architecture by May 2008

Sign In