Matt, Bob, and I have made some serious progress this week on the OFDM
code
and are close to a fully running transceiver. The attached figure shows
how
we equalize the channel based on estimates from known symbols at the
start
of the frame.
The figure in the top left is the inverse of the transfer function of
the
multipath channel and the figure on the bottom is the equalizer
response.
They line up just so nicely that we couldn’t help but show everyone.
Matt and Bob have left, but we now have the simulation system working.
It
does fine and coarse frequency correction, phase locking, frame sync,
and
symbol sync. It did not work over the air, but we didn’t spend any time
looking into why; this will be happening soon. As soon as we see packets
over the air, all of this will be merged into the trunk.
Tom