RE: [Madwifi-devel] Per-packet TPC

[Henry Q. <he...@am...>]

> -----Original Message-----
> From: Neal Gokli [mailto:go...@st...]=20
> Sent: Sunday, November 09, 2003 9:45 PM
> To: mad...@li...
> Subject: [Madwifi-devel] Per-packet TPC
>=20
>=20
>=20
>=20
> On Mon, 20 Oct 2003 Sam Leffler wrote:
>=20
> > > On Monday 20 October 2003 03:37 am, Anthony Wright wrote:
>=20
> <snip>
>=20
> > > hardware. Also the ability to be able to do per packet=20
> transmission=20
> > > power, so if you're talking to two people, one close and one far=20
> > > away you could be sociable and lower the transmission power when=20
> > > talking to the one who's close.
>=20
> > Per-packet TPC is possible on the 5212 and later parts but=20
> isn't yet=20
> > available (even for Atheros drivers).  It will happen together with=20
> > support for hardware multi-rate retranismission (another=20
> 5212-specific=20
> > feature).  Right now the main focus is getting support in for a new=20
> > part that has already hit the retail channels and which,=20
> > unfortunately, the existing HAL code doesn't handle=20
> properly (instead=20
> > of rejecting the part as unsupported it attaches and then jumps=20
> > through a null function pointer--oops).
> >
> > 	Sam
>=20
>=20
> First of all, thanks to all the developers for the great=20
> work.  A lot of us really appreciate it.
>=20
> I have a question about per-packet transmission power=20
> control.  A lot of research is going into new transmission=20
> power control schemes, but nearly all of these have gone=20
> untested in the real world.  The hardware has not supported=20
> it until now, and it has been difficult to find out if=20
> limitations are in hardware or in the drivers.
>=20
> If 5212 supports this, a lot of people in the research=20
> community will be happy to hear about it.  I plan to make an=20
> easy interface for researchers to test their TPC schemes, so=20
> they can validate the results of their simulations in the=20
> real world.  Can I just change the txpower parameter of the=20
> ath_hal_setuptxdesc() call for each packet that is sent?

It needs a HAL change in order for that to work.

  If=20
> so, how do I find what values are acceptible?
>=20
> Thank you much,
>=20
> Neal

Henry
>=20
>=20
>=20
> -------------------------------------------------------
> This SF.Net email sponsored by: ApacheCon 2003,
> 16-19 November in Las Vegas. Learn firsthand the latest=20
> developments in Apache, PHP, Perl, XML, Java, MySQL, WebDAV,=20
> and more! http://www.apachecon.com/=20
> _______________________________________________
> Madwifi-devel mailing list
> Mad...@li...
> https://lists.sourceforge.net/lists/listinfo/madwifi-devel
>=20