Silicon Investor (SI) -- The First Internet Community

STOCKTALK

We've detected that you're using an ad content blocking browser plug-in or feature. Ads provide a critical source of revenue to the continued operation of Silicon Investor. We ask that you disable ad blocking while on Silicon Investor in the best interests of our community. If you are not using an ad blocker but are still receiving this message, make sure your browser's tracking protection is set to the 'standard' level.

Technology Stocks : LAST MILE TECHNOLOGIES - Let's Discuss Them Here -- Ignore unavailable to you. Want to Upgrade?

To: Frank A. Coluccio who wrote (3148)	3/17/1999 11:07:00 AM
From: WTC	Read Replies (2) \| Respond to of 12823

ADSL causing interference into T1s? I'm dubious, but this is not a simple or trivial question -- a lot of smart, experienced ILEC staff worked many man-years on the hypothesis over the last couple of years to become satisfied that it would not be a problem, before wide scale ADSL deployment. You will recall that was the period when the ILECs were assailed for dragging their feet and creating xDSL red herrings by imagining "unlikely" interference problems into existing services. It is impossible to critique Powell's hypothesis without more definition of the problem that Powell cites. Is he talking about T1s on repeatered lines, with AMI line code? If so, I am very dubious that aggregate mutually coupled energy from ADSL lines in the binder group or sheath could impact the T1 error rate. Frank's analysis is valid, though -- if for some exogenous reason, any T1 is operating with virtually no transmission margin -- at the edge of the abyss, so to speak, then we have the butterfly flapping his wings from chaos theory. Anything might throw the T1 into error bursts, including temperature changes, or slight voltage changes in the line powering. The more interesting hypothesis is that the aggregate energy from ADSL services might impact HDSL services. In this case, though, I believe Covad's Pelosi is right -- the 2B1Q line code has been shown to be damn bulletproof in the electrical influence of CAP and DMT line code ADSL services. The parts of Powell's contention that I seriously question are his CAP vs. DMT generalizations for range and interference potential. I have not seen paper analyses or experimental data suggesting that DMT causes less interference into T1 than CAP. I have definitely not seen any suggestion that CAP has longer range or a higher percent of households servable in twisted pair plant than DMT, in fact, we found the opposite is slightly true. And that comes from a company that has deployed CAP and is migrating to DMT -- first using CAP for market entry timing reasons, not technical reasons. Powell's asserts that the interference is happening, that the ILECs just haven't noticed it yet because it hasn't shown up in large amounts yet. I assume he envisions a slowdown in throughput because of increases in burst/bit errors and frame re-tranmissions. If so, that is a pretty conspicuous situation -- elevated BER. If this is into an old-style AMI T1 line, there would be no place to run, since capacity is generally engineered on a binder group basis. With HDSL designs, unless you move to a different cable sheath or add HDSL regeneration, I am dubious that a pattern of interference could emerge that might be effectively masked by "repair" through pair changouts. The larger reason for my doubting Powell's hypothesis, though, is 1) the testing that has been done on xDSL NEXT and FEXT interference parameters, and the guidelines for successful deployment that derived from such testing, and 2) the historic, oft damned, but perhaps useful (here) tendency of ILECs to engineer very conservatively. I doubt if any ILEC is deploying their xDSL to distances right at the theoretical interference limit. I know that at least one is not.