Mobile data congestion will force a revolution

Printer-friendly version

Mobile phone operators are becoming increasingly anxious that they will not be able to fulfill customer’s expectations in terms of mobile Internet access. And they are right. Thanks to the emergence of high performance mobile devices and the coming of age of a 100% mobile generation, Cisco expects Western Europe's  mobile data traffic to increase 37 folds between 2009 and 2014 to more than 1bn Gigabytes per month.

This looks huge, yet it will represent only 7% of the total Wester European  IP traffic (Europe is more data mobile than average, the proportion being 4% globally) .Since few observers expect the Internet to collapse in the coming years, the logical conclusion is that we have a mobile specific issue. This is correct. The problem is spectrum.

Unlike fixed networks, where it is always possible to put more fibers along the existing ones, radio frequencies are natural resources and come only in limited supply. We have pretty much exhausted the current supply. True, the phasing out of analogue TV will provide some relief but nothing capable of matching such an explosive demand. Some improvement is also expected from the technology (new codes, better compression, faster hardware and so on) but these too are too limited to solve the issue unless an (unlikely) spectacular new technology emerges.

The only solution is to decrease the cell size: As the cell radius goes down, the number of active devices per cell decreases and the available capacity per device increases. This solution has other benefits: As distance is shorter, needed transmitting power is lower, batteries last longer, and electromagnetic pollution decreases. The big drawback is that as the radius goes down, the number of cells (and thus base stations) needed to cover an area increases exponentially (when the radius decreases by 3, the number of base stations multiplies by nine).The challenge, if one really wants to meet the demand is thus to deploy millions of very small cells (called femtocells). Clearly this cannot be done by any operator.

The only solution is to use the existing base of private Wi-Fi base stations when mobile consumers are static and to keep 3G/4G networks when they really move.True this creates a number of difficulties, namely in billing, revenue sharing, roaming, security and so on. But there is little choice. Operators are currently revising their data offers and everybody is watching ATT with interest who decided to close its unlimited offer. This approach is a dead end. When the demand has increased 37 times, the current limit of 2 gigabytes per month will look way too small. Operators will have no choice but raise the limit very significantly or see consumers go away.

History shows that when telecom starts becoming a problem, IT users find innovative ways to circumvent them. Internet started exactly with the idea of hitch-hiking unused capacity to circumvent expensive fixed line. If mobile congestion becomes a problem and mobile data becomes too expensive, it is only a matter of time that somebody invents a way of sharing Wi-Fi capacity on a large scale. If operators do not move, the revolution will happen in their back and they will end up in the same situation as the fixed operators when the Internet emerged..