| Thread Links | Date Links | ||||
|---|---|---|---|---|---|
| Thread Prev | Thread Next | Thread Index | Date Prev | Date Next | Date Index |
|
Attendees:
Dong Wei (person from Ikanos) (person from ST) Bernard Debbasch Tai-Lai Tung Michael Beck Barry O’Mahony
Michael prepared new text for Table 62B-1, based on last week’s discussions, which may be found at http://www.ieee802.org/3/efm/public/nov03/copper/62Bd2_2CMP.pdf
Attendance was light for this call. No one called in from Geneva; presumably activities in SG15 made that impossible.
However, several email exchanges occurred immediately prior to, and during, the conference call. Opinions in them served to form a basis of discussion during the call:
-- UPBO can significantly complicate test cases. This is clear from the results of Vladimir’s contribution. Proper TX PSD must be verified in each case with UPBO on. Many opinions expressed that only a few tests with UPBO on would be needed to verify proper functionality of the mechanism. The remainder of the tests, in particular those with AWGN-only and Self noise models, do not in fact need to be tested with UPBO (ETSI and T1 do not define cases in this configuration).
-- Michael will produce new text for 62B-1 with the following changes:
- UPBO will be enabled for Tests 2-9; that is, tests with A or F noise models. All other test cases will have UPBO off.
- Tests 2-9 will also add 20-Self crosstalkers to the noise model (together with the existing AWGN and either noise A or F). Self crosstalkers will be set to same loop length and UPBO PSD reduction as the units under test.
A comment may be needed against Clause 45 if additional control is needed in order to “turn off” UPBO.
There were no proposals received at this meeting for assumptions to input into the software simulator.
Next meeting is tentatively scheduled for Tuesday, 28 October, at 18:00 Geneva time (9 a.m. PST).
|