Re: [10GBT] Discussions on precoder decisions

[Gottfried Ungerboeck <g.ungerboeck@BLUEWIN.CH>]

George,

 

the 128DSQ constellation ("double square") is the natural go-in-between 8 PAM and 16 PAM, much more so than 12PAM. Contrary to 12PAM, all encoding, decoding, and precoding operations are based on powers-of-two arithmetic. 128DSQ has been known for a long time in the literature and its good fit with precoding has been well described.

 

On precoding, please see ungerboeck_1_0704.pdf (Portland). One fixed IIR-type precoder is enough. If the group cannot agree on a single fixed precoder, then let it be a very small set (2?) of fixed precoders, one being selected preferrably during auto negotiation.

 

Regards, Gottfried

---

From: stds-802-3-10gbt@IEEE.ORG [mailto:stds-802-3-10gbt@IEEE.ORG] On Behalf Of George Eisler
Sent: Mittwoch, 15. September 2004 06:03
To: STDS-802-3-10GBT@listserv.ieee.org
Subject: Re: [10GBT] Discussions on precoder decisions

Hugh,

 

Thanks again for taking the badly needed lead on urging along the decision making. The precoder definition is clearly in need of consideration, although I think it will take simulation work to really nail it down. Of course you can't finish it if you don't get started.

 

On last week's call, I was surprised by the "double square" contribution, with it's sort of 16/8 PAM flavor. It seems to open the case up rather than converging, but I wonder what's your take on it? I was going to call you, only to realize that I don't have your phone contact.

 

George Eisler

310 459-9225