FOSDEM attendee? Re: 802.15.4 CSS PHY

GNU Radio
1

Hi list,

this is just a shot in the dark, but I forgot to get the name of an
attendee during the Internet of #allthethings talk.

If you are, or know who is, the person who said he had implemented an
802.15.4 CSS PHY in the question period, please get in touch with me
or pass on the message.

It would be very beneficial to collect the 15.4 related work into one
repository to reuse some amount of code. The 800 & 900 MHz PHY would
be very useful to have in there and would be relatively painless to
design. The CSS PHY is obviously much more interesting :slight_smile:

Also, in particular, any future UWB PHY designers would be helpful.
The SDR hardware for such a PHY might be a bit pricey. Even a
unit-test driven PHY design would be helpful or simulation.

As of this point, I have not heard of any major silicon vendors are
even speaking about offering an UWB PHY, so it would be nice to be
ahead of the game.

Cheers,

C

2

Hi,

that was me ;). I also talked to Bastian B. and we agreed that I
will
create a pull request to gr-ieee802-15-4 once I finish my master’s
thesis
and have it all cleaned up. In the meantime, you can find my fork of
Bastian’s repo here: GitHub - fwunsch/gr-ieee802-15-4: An IEEE 802.15.4 ZigBee Transceiver.. The branch
with my work is called css_phy. Currently I’m working on the
synchronization for the CSS PHY. My previous work focused on evaluating
PHY
and MAC performance with the assumption of perfect sync. It’s also not
really complete yet (e.g. preamble and PHR are dropped in the receiver
because their content is assumed to be known) and endianness might be
wrong
at some points. But I will try to fix that in the next weeks.

Feel free to contact me if you have any questions!

Greetings

Felix

2015-02-02 7:52 GMT+01:00 Christopher F. [email protected]:

3

Hi Christopher,

in your talk you mentioned some bugs that you found. Would be great if
you could create some patches and a make pull request.

If someone is interested in the 800 MHz phy I will look for cheap HW and
try to implement them.

Best,
Bastian

On 2015-02-02 15:39, Felix W. wrote:

receiver because their content is assumed to be known) and endianness

repository to reuse some amount of code. The 800 & 900 MHz PHY would

Discuss-gnuradio mailing list
[email protected]
Discuss-gnuradio Info Page

–
Dipl.-Inform. Bastian B.
Distributed Embedded Systems
University of Paderborn, Germany

4

Excellent!

Thanks for your reply Felix!

As Thomas R. pointed out, the sync problem would ideally be solved
with the correlate and sync block, as 802.15.4 packets have a known
preamble and SFD but there might be some glitches with it. However,
Thomas
also optimized Costas Loop to be quite efficient now (see [1]). I’m
doing a
bit of experimenting on synchronization for this week and will submit a
pull request to Bastian as well.

Cheers,

C

[1]

©

Sent from my Android

5

Will do!

©

Sent from my Android