I realize that there is probably a whole lot of wireless traffic, and that it is increasing. But to presume that Moores law will apply to something that is limited by physics is a bit of a stretch. It did apply to manufacturing, which is a bit different. My feeling is that these "market predictions" are actually what the marketing people want it to be, rather than valid numbers based on any reality. Besides that, there is only so much spectrum, they are not making spectrum any more. Yes, we can always go to a higher frequency, but physics makes that more difficult as well.
@dirk.bruere: the number seems a bit small to me as well for wireless traffic. The wired traffic is already many orders of magnitude higher (elsewhere on EE Times forum I posted traffic data in major internet exchanges already crossing TB/sec).
It is clear to me that the content demanded by smart phones and tablets will be more data-intensive as the world adopts more of 4G-wireless services. The data traffic by M2M nodes will be far less and is not the major demand for increased data traffic.
Moore's law will be required to be applied to the bandwidth usage now, what he has said for the usage of transistors in the ICs is going to be true for bandwidth usage in wireless.
In my opinion Agilent is having quite right step further in formulation of strategy.
Blog Make a Frequency Plan Tom Burke 17 comments When designing a printed circuit board, you should develop a frequency plan, something that can be easily overlooked. A frequency plan should be one of your first steps ...