[Discuss-gnuradio] WiMAX decoding: DL-PUSC question

[Alexander Chemeris]

Hi all,

This is a little bit offtopic for this mailing list, but I can't find
a better place to ask this question.

We're doing an open-source project on WiMAX decoding
(http://code.google.com/p/wimax-scanner/) and at this point we have a
trouble with correct subchannelization of DL-PUSC, so we're looking
for some advise on what we're doing wrong. Or may someone could just
contribute some code for correct subchannelization?

My implementation is available as Matlab code and as Excel spreadsheet here:
http://code.google.com/p/wimax-scanner/source/browse/matlab/get_slot_data.m
http://code.google.com/p/wimax-scanner/source/browse/matlab/PUSC-permutation.xls
It is able to correctly list sub-carriers for FCH of segment 0 (i.e.
for subchannels 0-3), but looks like it gives incorrect results for
all other subchannels. The result of this is that I can see only two
consecutive repetitions of slot data in DL-MAP part of the header,
where I should see 4 or 6 repetitions. The standard is very vague and
unclear on this topic and people who write textbooks certainly thinks
it's an obvious part and no one document it in an understandable way.

If you have a working implementation of DL-PUSC subchannelization,
which you can't share with us, we would appreciate if you just share
mappings of subchannels to sub-carriers for a few sets of parameters.
That would be very helpful for us to debug our own code.

PS If you're interested in contributing to an open-source WiMAX
implementation - you're very welcome to join the effort! We believe we
can make a real difference in the wireless world.

-- 
Regards,
Alexander Chemeris.
http://www.fairwaves.ru