revisiting the 434 MHz band

Werner Almesberger werner at
Fri Jul 23 07:53:02 EDT 2010

Wolfgang Spraul wrote:
> OK so you are basically saying the SiGe stuff is a bit too low-level :-)


> I'll try to understand the MAX2830 you pointed out, if you find out more
> please keep sharing...

I think it's still too low-level for our purposes. That is, unless
there's some nice chip around that picks things up at the analog
I/Q level and produces at least the bit stream. (And vice versa in
the transmit direction.) From there on, a CPLD/FPGA should be able
to do the rest without too much suffering.

There's also some very recent (March 2010) material available for
GNURadio and the USRP2 (802.11p, but includes 802.11a/g as well.
802.11b has already been done earlier):

This work could at least serve as a guideline for what needs
implementing, plus a testbed to debug an implementation.

Now the question is whether such a "missing link" chip can be
found. It would have to contain ADC, DAC, and signal processing.
With discrete ADC, DAC, and DSP, or a USRP-like design that just
sends everything to a general-purpose CPU for processing, you'd
end up with an uncomfortably large number of components.

I looked at Maxim's Web site for hints of what they would
recommend to connect to their chip, but they seem to be content
with leaving this to the customer to figure out. Also their eval
boards just bring out the analog signals.

I'm not sure 802.11b/g is a realistic target for the Ya with this
sort of approach. I'd see a much better chance for getting things
right in time with IEEE 802.15.4. Of course, IEEE 802.15.4 has
its drawbacks, there being lack of ubiquitity for the Internet
access we all crave, possibly short distance (to be verified), and
lack of Open-friendly and directly useful protocols to layer on
top of IEEE 802.15.4.

A third option would be BT. Range should be similar to IEEE
802.15.4 or better, the protocol suite is widely implemented, but
it still doesn't give easy public Internet access, and I don't
know about openness. I also don't know if there are any chips
that offer a degree of integration similar to what's available for
IEEE 802.15.4.

- Werner

More information about the discussion mailing list