Blog Post

Can Personal Cellular Sites boost cell service?

By Allan Leinwand 

Mobile operators are on the verge of asking you to help them solve one of their biggest problems – how to get more signal strength where you need or want it most. Their plan? Allow end users to buy personal devices that act like Wi-Fi routers, providing nearby cellular bandwidth in hard-to-reach places like offices and homes.

These next type of cell sites, named femto cellular (femto being smaller than pico, the term used by mobile operators that refers to smaller cell sites) are setting out to solve carriers’ often-expensive problem of providing complete coverage. Mobile phones usually work well in metropolitan areas, but travel a few miles off the Interstate or into the country and signal bars drop rapidly. Most frustrating to many people is that the signal strength at their homes or inside offices is often unusable.

The forthcoming femto solution? Having end-users buy a small femto device, similar in concept to a Wi-Fi access point, that is a personal cellular site. The femto cellular device has a cellular antenna to boost the available signal as well as an Internet connection. The device uses your Internet connection to connect to your mobile provider’s’ network and route your phone calls.

There are a few limitations, or benefits, to this approach, depending how you see it. First of all, the femto device you buy will probably only connect to a single mobile provider’s network. That’s good if you like your mobile operator and bad if you want to switch operators on a regular basis. This approach is clearly good for the mobile operator because you buy a device that uses your Internet connection to extend their network and gives you less incentive to switch providers.

Since femto cellular devices are not available yet, there are some unknown issues – will mobile operators charge the same for minutes via femto cellular devices? Will enterprises buy femto cellular devices like Wi-Fi access points to extend cellular coverage? How do you stop your neighbors from using your femto cellular device and the associated broadband bandwidth (or do you care)? And how much are you willing to pay for a device that lets you use mobile phones in your house?

Allan Leinwand is a venture partner with Panorama Capital and founder of Vyatta. He was also the CTO of Digital Island. 

26 Responses to “Can Personal Cellular Sites boost cell service?”

  1. Jack Reacher

    What about the ISP? How are they going to feel about all this? If my ISP is AT&T and my cell service with Verizon….is AT&T going to let my VZW calls go through their network?


  2. IP.access has already developed a so called “femto cell” – a WiFi-router-sized box (120 mm × 162 mm× 40 mm), that provides 3G/HSDPA coverage within a distance of up to 200 meters. The device operates in the licensed 2,1-GHz-band at 5 mW and allows HSDPA (16QAM) connections with 7,2 MBit/s besides voice telephony. The uplink will be established by any broadband internet connection through an ethernet port. Seamless handovers from and to regular 3G-cells are to work, too.
    ABI research expects more than 32 million of such or similar devices to be installed by 2011, which will then serve more than 100 million users.

  3. Jesse Kopelman

    DeadCellZones, a femtocell is neither a WiFi device nor a repeater. It is just a very low power and capacity cell site — conceptually no different from what’s on a tower, just much much smaller. The device would still be provisioned and controlled by the carrier and would need an always-on style connection (but no need for broadband speeds) back to that carrier’s switch. I would be very surprised if there is major support for this technology from major US carriers — it is just too different from their SOP.

  4. Good concept, bad idea.

    I’m sure a decent proportion of people that live where there isn’t a cellphone singnal probably have a dialup provider. The last thing I want is to call someone and have it sound like I’m talking to someone who is both on a cellphone and using Skype over dialup.

  5. Jesse Kopelman

    This is not a new idea and the technology has been available for years. In theory, you don’t even need broadband, as you can avtually fit a couple of voice channels over a good dialup connection. The big issues are these: cost and control. How do you get people unwilling to get anything other than “free” phones to shell out $500 for a femto/picocell? If you aren’t selling millions of these things, the price may not even get as low as $500 per and certainly won’t get lower anytime soon. As for control . . . well the way most carriers feel about their spectrum is pretty similar to the way Tony Montana felt about his sister. The big thing femto/picocells have against repeaters is reduced impact on the surrounding network when “improperly” installed. Still, the added complexity is huge and with complexity comes price.

  6. GSM codex can compensate for a bit of latency. They can also squeeze allot of calls into a small space. The initial deployments fit all the calls from 3 TRU’s (transceiver units) into 3 timeslots on a T1. They are squeezing even more calls into 3 timeslots now. There are 24 timeslots on a T1. Each 64Kb timeslot is handling anywhere from 7 to 14 people.

    Security of course is also an issue. GSM uses encryption on the RF side but is open to various test equipment (siemens and others). For about $1M, you can buy a device to listen to GSM traffic.

    On the IP side, I am sure they must encrypt, otherwise people in the US would reverse engineer the traffic in no time. Ill ask.

  7. Are you kidding me? This would potentially open handsets to huge security exploits. As soon as one guy hacks his picocell box and starts sending OTA data to his cell to reprogram it, it’s all over.

    The providers will never go for it.

  8. Allan Leinwand

    A few coments…

    James Hancock: Femto cellular is different from a cellular repeater. Femto is a cell site itself and connects back to the provider’s network using Internet bandwidth and an IP tunnel. A cellular repeater simply repeats signal that is present to make it stronger – no Internet involved. You’re right that both solutions can solve the problem. Yet, I think the femto model has some merits because you don’t need cellular signal present for it to operate.

    SG: You are right that latency will be an issue. A rule I have been told over the year is that if latency is less than 200ms or so then voice quality should acceptable to the ear. Since you can generally route a IP packet across the country in less than 80ms, then latency may not be an issue. Of course, last mile and backbone congestion, your location in relation to a major Internet hub and packets crossing an ocean may affect your latency dramatically.

  9. Perhaps these mesh WiFi vendors had the intelligence to have their devices have RF reception for 800/900/1900 MHz so that along with the 2.4/5 GHz for 802.11 a/b/g it could also receive Mobile phone signals. Then al they have to do is take the digital phone signals and have it routed over the IP network to a central location where it can be converted and emitted out as RF.

    Would be interesting to see what it does to the latency and the voice quality.

    Martin, I just you the next Fon device idea. Maybe the Meraki folks could also use this for their next product.

    AS a matter of fact I would love to try this out myself at home myself. Maybe a new startup.

    Any feedback, please shoot me responses at my mentioned email address.

    [email protected]


  10. I have a non-existant cell phone signal at my home. A little problem called Hills. Anyway I am looking around at various options for boosting the signal. Given I can see a cell phone tower from my house I would have thought I should have a signal. Anyway a good and timely article.

  11. You can already do this with a myriad of repeaters that are on the market. You don’t need to use a specific carrier and they work quite well. Just pick the band (or bands!) that you want to work with (800/1900 etc) and volia you’ve got fantastic cell service. Ones for houses are < $400.00 right now. Malls use these all of the time and those run

  12. This isn’t really new, and is already in use, with the UMA phones in the UK and France. Or am I missing something here?
    There it’s coupled to a lower tariff with calling from home.

  13. Perhaps if carriers had focused more on, say, building more sites instead of trying to make us watch TV on our phones or buy $1.99 ringtones that expire in 90 days, we wouldn’t have this problem.

  14. Matt Hocker

    I have another thought – what about using these femtocells as a way to bypass roaming charges? Theoretically, you could plug one of them into a hotel broadband connection or at a second home in a lush island paradise and carry your home rate plan with you, just like is possible with VoIP now.

    I would expect that this would be dramatically illegal. Using licensed spectrum without permission – especially expensive GSM spectrum – is generally frowned upon. It could even cause technical problems to existing carriers. For example, what if you plugged in a GSM femtocell in Korea where there is no GSM, just CDMA – would it cause interference to the CDMA networks?

    Still, this is an extraordinarily disruptive proposition to the wireless model and might just be the reason that the (over-protective, conservative and generally clueless) wireless carriers won’t do it.

    I would buy it in a heartbeat, though.