The boss of TI's wireless operation Greg Delagi, sees an almost infinitely extendable future for mobile phones as they morph into 'smart mobile-companions'.
In his keynote presentation to next February's ISSCC 2010 in
Their displays will have 3D imaging, virtual interaction and conferencing which make every call 'feel like you are right there'.
They will be 'always-on always-with automatic choice of the most appropriate modem link, whether WiFi, LTE, 5G, or mmWave, based on which will best deliver performance, battery life, and costs to deliver the best-possible user experience.
They will have all-day operation or two-day always-on standby without a recharge.
They will use innovations ranging from low-voltage digital and analogue ICs, to non-adaptive power management and energy harvesting, to extend battery life to a week or more.
They will combine increased bandwidth with decreased latency; lower power requirement with energy scavenging and harvesting; multimedia processing power with new interface technologies.
... and most of us will still use them just to make phone calls :-)
How very true, Mike
...but you still won't be able to get better than GPRS data rate except in big cities.
Sadly, Keith, that's only too true. The network operators make the plans of the hardware developers look absurd.
The problem is that network capacity is limited by Mr. Shannon, who is not as soft a touch as Mr. Moore :-)
Right now it's easy to double the demand for mobile data bandwidth every year or so by building faster more data-hungry mobile devices and then selling more of them.
It's impossible to keep up with this demand by cleverer modulation schemes any more, we're already close to the Shannon limit. So even with new radio systems (LTE, WiMAX, 5G...) the only solution is to build more basestations -- and quite apart from the NIMBY problem of where to put them, doubling the number of basestations is a huge (and very expensive) task.
And then you'd have to double them again a year later, then again the year after that...
Ian
What an awful thing to think, Ian, that the mobile datacoms experience is unlikely to get any better.
Ian: which is why some people are pushing for picocells. But those still cost a lot of money/time for getting the backhaul to them, so others believe femtocells are the only solution. Too bad asking users to pay for their own femto then make it publicly accessible by anyone in the street is completely insane and a genuine rip-off, so the business model to make it work hasn't been developed yet.
The notion that a wireless network operator develops the network will need to go away in the long-term. How the initial investment is paid back and to who is the big question. Could we get phones/data devices to instantly pay for any bandwidth they use to the individual owner of a femtocell? Seems like a security nightmare to me, but intriguing otherwise.
But the classic way to get more bandwidth is to go up in frequency, and people are already talking about 60 GHz for future mobile systems: now there's a challenge!