Wi-Fi's likely impact on 3G/4G services and operators

This morning at the 4G Wireless Evolution conference in Miami, I gave a talk about how Wi-Fi is going to impact both 3G/4G operators and fixed line operators over the next 2-4 years.  The slides are on SlideShare, and here:

Wi-Fi Opportunities In A 4G World

View more documents from Brough Turner.

I think the reason I’m invited back is I manage to be controversial and since, today, I was given more than an hour all for myself, I attempted to make at least a few provocative points:

• We’re at a wireless tipping point that will dramatically drive up performance of all wireless systems, but Wi-Fi is way ahead of WiMAX or 3G/4G mobile, thus Wi-Fi is where the excitement will be over the next 2-5 years.

• Femtocells will flop.  They do provide a way to extend voice coverage into homes that macro cells don’t reach, but they are not efficient for data offload.  Since Wi-Fi is efficient for data offload, and it costs less to buy and less to operate, Wi-Fi is will trump Femtocells.

• Wi-Fi is consistently ahead of the 4G community in commercial deployments of “4G technologies.”  Specific 4G technologies that Wi-Fi has pioneered include OFDM modulation (802.11a, 802.11g, 802.11n), MIMO (802.11n) and beam forming (done via antenna element selection as early as 2002;  and done via antenna arrays with on-silicon signal processing under 802.11n, coming to market right now).

• MIMO makes 5 GHz spectrum as useful as 2.4 GHz spectrum or TV spectrum. 

• Beam forming via adaptive antenna arrays will dramatically drive up performance (for Wi-Fi beginning now and for WiMAX and LTE some years from now).  Beam forming increases range, reduces interference and allows spectrum reuse even in confined areas.

• But with beam forming 5 GHz spectrum becomes more valuable than 2.4 GHz or TV spectrum as shorter wavelengths (at higher frequencies) allow tighter beams.

• As Wi-Fi performance and range increase, it’s becoming useful for broadband wireless service delivery, thus driving down costs and helping fuel rapid growth in the Wireless ISP community.  (There are over 2000 WISPs in the US today.)

In my talk, I backed up these statements with data and arguments that may not be clear from the slides alone.  If there is any point you don’t agree with or don’t understand, fire away in the comments below and I’ll endeavor to answer within a day or two, or elaborate in a separate blog post.

Comments

[Alan Little]

What you say may well be technically true. But business models and billing will be the killer.

I too thought, maybe five or six years ago, that Wi-Fi's ubiquity in places where business travellers want to be connected - airports, hotels, railway stations, conference centres - would knock a big hole out of the business case for then-still-expensive UMTS data. What actually happened, though, was that UMTS data plans became reasonably priced, and UMTS/HxDPA became fast enough, cheap enough and more or less ubiquitous. Whereas to use Wi-Fi I have to go through irritating billig gateway pages with numerous silly little ISPs at extortionate hourly rates, and the performance gain over HSDPA just ain't worth it. I think in the couple of years since I had an HSDPA- and tethering-capable handset, I have paid for Wi-Fi service exactly once.

The sensible option - for customers as well as for operators, since it could potentially offload a lot of traffic from expensive UMTS networks - would be for operators to bundle Wi-Fi and UMTS connection in a single data offering. But here in Germany, I believe only T-Mobile does. I don't know why other operators (including the one I work for) don't.

[brough]

Alan, Unfortunately, the slides alone don't convey what that actual presentation included. Sorry, I don't have a recording of the voice track. :(

I agree with your last paragraph. I was making the point that not only should this happen, it will and it is already here in the US.

What I said was Wi-Fi is going to show up as a "free" service in the "freemium" sense, i.e. it's free because it is sponsored by your 3G/4G carrier or your fixed carrier. Wi-Fi infrastructure is much less expensive to deploy, at least in urban environments, and (depending on backhaul arrangements) much higher capacity.

Already in the US we have Cablevision bundling "free" Wi-Fi access for the Internet access subscribers on Long Island and southern Connecticut. To do this, they deployed over 20K outdoor Wi-Fi access points with DOCSIS backhaul over their cable infrastructure. In response, Verizon FiOS has struck a deal with Boingo to provide free access to Boingo's many Wi-Fi access points if you are a Verizon FiOS Internet access subscriber. And this in a market that has only a duopoly.

Of course it's the 3G operators who have capacity problems and need Wi-Fi offload. And, sure enough, AT&T Wireless has a program to make AT&T Wi-Fi hot spots available "free" to anyone who has an AT&T data plan. Note there are over 20K AT&T Wi-Fi hotspots already deployed.

Finally, many many locations are going to "free," i.e. sponsored, access. Even Logan airport here in Boston is opening up the formerly expensive Wi-Fi service and making it free.

I don't like Wi-Fi billing screens or the ludicrous charges that I encounter in airports and hotels. I think we're in agreement, operators should bundle Wi-Fi access and indeed, that just what is happening here in the US.