OLPC hardware: what if there was an SDR modem / chipset?

[Luke Kenneth Casson Leighton]

On Mon, Jan 25, 2010 at 6:35 PM,  <david at lang.hm> wrote:

> [...] projects. I'll bet that these chips can operate in that band, but not
> throughout the band without changing external components.

 certainly mirics 80mhz->5ghz receiver chip has multiple RF inputs and
has i think it looks like it has 3 separate RF down-converters all
capable of running simultaneously, so it could do e.g. FM Radio and
DVB-T at the same time.  whilst i don't have enough info at this stage
to confirm whether that's the case in the two transceiver chips i _do_
expect it to be so, simply because... well... the designers would have
to be daft not to have done it! :)

 so, i expect that the chips will be able to operate in several wildly
different frequencies simultaneously, _if_ set up with several "sets"
of appropriate external components (with matching antenna) but again:
too early to tell, yet; i'll have more info later as my research
progresses.

 the point is, however, to establish the actual cost of a
multi-frequency SDR modem in OLPC-level mass volume, and see if it's
viable.

 _orrr_.... establish whether the next gen OLPC hardware could put in
a standard interface such as PCI-e (USB2 not SATA/PATA) which could
easily be used to slot in an Intel 5350 WIMAX (or even pffh one of the
PCI-e 3G modems from Huawei such as the EM775 or EM770); that would
then be "future-proof" upgradeable, so that if a multi-frequency SDR
modem ever came out in 50x30mm or 24x30mm mini PCI-e form-factor, it
could just be slotted in without a PCB redesign.

worth thinking about?

l.