I’ve used wireless 3G services since 2004, and I’ve seen the experience degrade over time as demand for these networks has outstripped supply. More people have discovered the joys of using the Internet everywhere, which has led to growth in adoption and more devices with embedded 3G radios. In turn, carriers are unable to balance the growing demand with their limited supply.
And it’s not just data traffic that’s causing problems — it’s all too common for voice calls to be dropped due by overwhelmed networks or limited geographic coverage. But without billions of dollars in additional infrastructure investment, carrier networks are limited to their fixed coverage areas.
In an effort to limit network expansion (and its associated costs), carriers are starting to push femtocells as a solution for poor coverage in the home. These router-sized devices are essentially miniature cellular base stations that use a pre-existing broadband connection for backhaul. The potential market ranges from home users to small businesses where people use a cellphone in lieu of a fixed landline. From the user perspective, femtocells mean more reliable phone service. From the carrier’s point of view, femtocells offer bits of targeted network expansion where it’s needed most, and at a relatively small cost.
But can fixed-base femtocells really rescue us from the plights of cellular constraints?
Just an update to this article — ABI has slashed its femtocell sales expectations for 2009 by 55% and for 2010 by 40%. Various reasons are involved — some I’ve touched on above, and some are just the result of the current economic climate.
Kevin,
I was enamoured by the Palm Pre and went to buy it on day one. On day two, I had to take it back because Sprint coverage at my home office was too spotty. When working from home, one cannot afford dropped calls. Two weeks later I heard about the Sprint AiRave. I was glad to purchase this device if it meant I could use the Palm Pre. All is well. In my humble opinion, Sprint should have provided me with this device for free and should not be charging me $5/month if they want me as a subscriber, however, all in all, I am grateful that this technology gives me more options.
Femto might be a big hit ones embedded into router/modem CPEs many households and small offices already use for spreading broadband internet connection.
Femot’s make tremendous sense if you’re a cable company rolling out a wireless service: coverage is the first issue. And yes – they should be provided like a free cable box…in fact is should be integrated with the STB. The perception with existing carriers (from dubious to outrage) offering femtos that use your broadband and then charging you for unlimted calling is justified.
Finally, don’t forget that coverage and capacity are related, not independent, in digital wireless adaptive modulation systems: a users in a weaker coverage area can be provided with more perceived bandwidth when there are less users on that cell/sector than during the busy hour.
Kevin,
I disagree with the majority of points in your article, but I’ll just focus on a few:
“Will these devices solve the underlying cellular network challenges of limited geographical coverage? Not by a long shot…billions in capital are ultimately needed to provide proper coverage and signal strength where people need it most.”
Well, you are mixing up the problem. The real problem, as you yourself stated earlier in this article, is one of network CAPACITY, not one of COVERAGE. Femtos are a marginal short-term benefit to the coverage problem, but are also a long-term powerful benefit to the capacity issue.
Second, do you remember the old stat that some 70% of phone calls are made when we are at our workplaces? Well, there is a vast amount of time that our cellular phones (and 3G laptops) are relatively stationary in our homes or offices. While there, our chatty smartphones are checking mail and moving data on a regular basis. This drone of useless traffic could be moved on to a femto. If carriers could move a good portion of home and office data use to the LAN, they could PERMANENTLY reduce the required capacity of the mobile infrastructure, saving the carrier on air-interface and backhaul capex and opex.
Your article makes scant mention of femtos in a workplace scenario, whether SME or larger. I’m not sure why you ignore that important usage case. Enterprise femtocells are now being offered that can handle more concurrent connections and offer wider range.
You didn’t mention that special calling plans could be offered to femto users when located in their femto network, such as cheaper international long distance calling, unlimited data, PBX-like services, etc. This could help push Fixed-Mobile Substitution. This is attractive to mobile carriers AND their customers, and has been proven mostly in the EU with “ZuHause”-type plans.
You talk about how Wi-Fi makes the femto redundant for mobile phones. I ask if the opposite might not also be true? It costs money to put WiFi in a phone, but is it necessary if we have a femto in the workplace and the home? Less so. Many subscribers only turn on WiFi for a couple of reasons: to drive up data speed, and to reduce cost if they don’t have an unlimited data plan. A femto could do exactly the same things. And WiFi is murder on a phone’s battery, but a nearby femto is actually good for battery life (as you wrote).
People with unlimited data plans and good coverage don’t have much incentive to turn on WiFi at the home or office. That’s fine for them, but puts a load on the carrier’s network. But if there were a femtocell, their traffic would shift off the WWAN automatically. Take, for example, an “iPhone family” – four people in a house with iPhones. AT&T should just send those guys a free femtocell “in appreciation of their loyal custoomers”, and get them off the WWAN! It’ll increase loyalty beyond the 24 month contract, and could pay for itself within the 24 mos.
Managing my phone’s WiFi (settings, WPA, WEP, turning off to save battery, etc.) is a pain, and I think the mass market would be happy to have none of it.
I like having WiFi in my phones, but that’s mostly for when I go abroad so I can use VoIP to cut roaming costs. That doesn’t apply to everyone. Most people stay domestic.
In conclusion, I think this article shows an unclear view on femtocells, their value proposition, on WiFi in phones, and on the distinction between coverage and capacity in a WWAN. If femtocells falter, it won’t be for the reasons cited here, it will be this main factor: cost. The cost of deploying and supporting femtocells needs to be cheaper than the comparable cost of increasing network capacity and coverage. We’ll have to wait to see which of us is right…
Derek, first of all, I want to thank you for the detailed and insightful commentary. It’s much appreciated, especially coming from someone like yourself that runs a telecom consulting business.
A few thoughts, with the first one being that I don’t think our viewpoints are as different as they might appear. I’m in vehement agreement with your points on the cost. I believe it to be a larger barrier to entry as mentioned in my article along with the direct link to how I feel the payment model is backwards. Your example of the iPhone family getting a free femtocell is spot on. However, I don’t believe that the customer should be charged any monthly fee either as they’re providing the backhaul and there’s a financial and network reduction benefit gained by the carrier.
I’d argue that the issue is one of capacity AND coverage, not just capacity, but perhaps that’s splitting hairs. In the end, the customers don’t care if their issues are due to a lack of capacity or a lack of coverage — they simply want the carrier to provide a solution for the paid service. As consumers increase mobility, I believe that coverage is a gap and a fixed femtocell can’t address that. Nor can Wi-Fi do so much better, but it’s far more prevalent in terms of locations, cheaper to implement and is growing in terms of coverage.
I also believe that customers DO have an incentive to enable Wi-Fi at home… in fact, I see two. One, it saves battery life in my opinion. I’d like to see some statistics and test results to better understand your “Wi-Fi can murder batteries” point. Two, it can be far faster as broadband speeds are increasing faster than mobile broadband speeds. For example, my 20Mbps FiOS line (both up and down) attached to my 802.11n router beats my 3G speeds hands down. And it will do so for some time to come. Yes WiMax, LTE and HSPA+ is on the way, but so is faster fiber to the home. That benefit is worth it to me to turn on my Wi-Fi at home… and it’s really not much of a challenge on my iPhone, Android or webOS devices. Two taps and I’m good. Maybe I’m the exception?
Your commentary about the enterprise opportunities is good. Again, I don’t own a telecom consulting company and as such, my focus is more attuned to consumer markets. I’m trying to change that and your response has helped me in that area. I appreciate the informative conversation!
Kevin,
Thanks for the reply.
You asked for info regarding “wi-fi can murder batteries”:
There is ample debate over whether Wi-Fi is good, or bad for battery life. The confusion stems from the fact that Wi-Fi has been steadily improving, and because it depends on the time-frame and the kind of activity the phone does in that time.
When actively sending data, Wi-Fi can save power over 3G. However, if left on during idle times, Wi-Fi saps power. For this reason, some phones (like PRE) automatically turn the Wi-Fi radio off when they lock. Essentially, cellular radios were designed from day one to conserve power, where wi-fi was designed with the assumption of steady power from the wall. So, to use wi-fi in a phone as a battery saver requires constant management of whether it is turned on or off. I think consumers don’t want that hassle, and would save power with it off – especially if they had a femtocell close-by which reduces the consumption of 3G power.
If you want to see a little research on the subject, there is some in this by Yuvraj Agarwal at UCSD:
http://mesl.ucsd.edu/yuvraj/appmaterial/Cell2Notify-Mobisys07_Yuvraj-Agarwal.pdf
Have a look at section 3.2, and table 2 therein. I’ll quote: “Even when the Wi-Fi radio is idle, the device consumes more than 15 times the battery power than in GSM idle mode.”
And regarding speeds, smartphones cannot handle data at the full speeds of Wi-Fi anyway. Websites don’t serve it that fast, either. Basically, the bottleneck in either a home Wi-Fi or Femto situation is not the last 20ft over-the-air hop. And Femtocells at shorter ranges, dedicating their full bandwidth to you, can get closer to the theoretical speed limits of cellular technologies than their shared-resource WAN counterparts. In short, the speed of either Wi-Fi or Femto will both be perceived as great.
Oh, BTW, I’m by no means focused on the enterprise market. I work across the entire telecom space, fixed, mobile, consumer, enterprise.
best,
Derek.
Good article Kevin. I agree with your view points.
With more and more devices that have embedded WiFi, I personally have a hard time figuring out the utility of FemtoCells. With Skype on iPhone for example or other VoIP/UC solutions, it is possible to make calls over WiFi today. Why is there a need for FemtoCells? I can uderstand the need perhaps for older cellphones which do not have WiFi…but moving forward, FemtoCells will be a hard sell. Solutions exist today that leverage WiFi for voice and of course data.
– Manju Mahishi
San Jose, CA
Thanks Manju. In chatting with Simon Saunders, I raised Wi-Fi as an option. Aside from what he said about interference and the inability guarantee service quality, he took the opposite view that you have. His thoughts are that Femtocells already work with every phone — be it a smartphone or a feature-phone — and that’s a key advantage of femtocells. I’m more inclined to agree with you on Wi-Fi since it’s proliferating into everything.