Let’s go out on  a limb and make a prediction. In five years, in dense urban areas, you will get your ADSL at cost, provided you subscribe to your telecommunications operator’s mobile offering.

Three major trends are at play :

  • Cells are getting smaller
  • Radio throughput is increasing
  • ADSL throughput is not going anywhere

Once cell throughput approaches ADSL throughput, the value of ADSL drops to zero. Why bother with ADSL when you have unlimited traffic at decent speeds with no geographical limitations ? In Paris, I seldom bother to even switch on my Android G2’s Wi-Fi networking – now it is all UMTS, all the time.

Why not ADSL for free ? As I can even get full motion video on demand on my mobile communicator, the availability of video services on the ADSL remains an incentive only if I’m interested in high definition. But to some people, high definition is important so ADSL retains some perceived value. In addition, giving away free ADSL access bundled with a mobile subscription would be gross abuse of a dominant position by operators protected behind the barrier to entry that their license affords them – so the worse they can do is selling ADSL at cost. But it is nevertheless tempting to squeeze out the perceived value of ADSL from the consumer’s point of view in order to cut the fixed access pure player’s oxygen supply. Isn’t life so much more comfortable among oligopolistic old pals ? Marginalization of ADSL pure players will be even worse if they are not playing along in the fiber optics arms race.

So the incentives combine :

  • Users want the convenience of permanent unlimited cell access
  • Operators are happy to squeeze out ADSL pure players

As a result, cell traffic increases and leads us to the next step of this self-reinforcing process : femtocells. Spectral efficiency nearing the Shannon limit, antenna diversity, spectral multiplexing and other 3G MIMO techniques can combine to provide the peak throughput that all the shiny marketing pie in the sky presentations promise. But in operations in the field those speeds are not achieved unless you camp under the antenna. For example, LTE 2×2 MIMO is advertised at a peak throughput of 173 Mb/s but actual rates are somewhere between 4 and 24 Mb/s in 2×20 MHz. They drop sharply as distance increases and it gets worse as the cell gets crowded. So there will be strong user demand for small cells – demand theoretically exists until there is one cell per user.

Approximately 60% of mobile usage already takes place indoors, yet providing in-building coverage is a technical problem at the gigahertz frequencies used for Wimax and LTE. This is only set to get worse as the mobile continues to replace the home phone. Research indicates that, as “all you can eat” data packages become commonplace, this number is likely to reach 75% by 2011.

Doug Puley – “The macrocell is dead, long live the network”, 2008

With the user spending more than 60% of his time indoors, there will be a fixed line access nearby. Extension of the access network on top of ADSL and FTTH links is already underway to increase capacity and compress costs by getting the data off the mobile network as close to the user as possible. Femtocells work well on ADSL too. So ADSL will remain useful as a way for mobile operator to shed load from the rest of the access network. And on top of that, ADSL lets the operator reach subscribers in areas not covered by the radio network.

So to mobile operators who offer fixed line access, ADSL could soon be considered as a mere adjunct to their core offering : mobile access. That could add yet more pressure on the game of musical chairs of mobile access frequencies license allocation. Why not attempt to exclude the competition that does not own a mobile network ? That leads us to ADSL access at cost – or slightly below that if the operator is willing to be naughty and deal with the market regulator. It will happen sooner than you think.

By the way, for a wealth of data about 3GPP evolution from UMTS-HSPA to LTE & 4G, you can take a look at this  September 2009 report by Rysavy Research. It provides about all you need to know about it and it is nearly as good as what I get internally from SFR.

Approximately 60% of mobile usage already takes place indoors, yet providing in-building coverage is a technical problem at the gigahertz frequencies used for Wimax and LTE. This is only set to get worse as the mobile continues to replace the home phone. Research indicates that, as “all you can eat” data packages become commonplace, this number is likely to reach 75% by 2011.