SI
SI
discoversearch

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 : PairGain Technologies -- Ignore unavailable to you. Want to Upgrade?


To: Jay Mowery who wrote (19456)1/27/1998 5:14:00 PM
From: John Stichnoth  Respond to of 36349
 
All: following was from Jan 1 issue of America's Network, a trade mag. (I found it while sitting with my 3 year old at her gymnastics lesson--you never know). Notable for who its written by--ADSL product manager for Analog Devises. It gives a clear explanation of plenty of issues, including those mentioned in the Aware announcements. Issue has two other articles on ADSL, but they are not up on their web site, and scanning these in is time consuming. I may try to put in excerpts if time permits.

What's next for xDSL ?
All hype aside~ xDSL keeps up the hectic pace
As ADSL grows more stable with maturity and a growing number of shipping products, an inevitable degree of differentiation will occur.

The digital subscriber line (xDSL) community is buzzing, with real developments and diversification almost keeping pace with the massive media hype surrounding them. Discussions are well advanced on Issue 2 of asymmetrical DSL (ADSL) T1.413. With the letter ballot complete, editing and reconciling the various comments are left before the new edition of the standard is published-which will be used as the baseline text within the International Telecommunications Union (ITU) study group
C H - C H - C HANG ES
Many of the changes has been to update the standard with the benefit of the last two years' experience. One specific change is to integrate asynchronous transfer mode (ATM) Transmission Convergence layer, so that the ADSL spec can be defined as a Utopia-Utopia connection, instead of a simple bit-pump. There are also some additions to the management and communications protocols to make these richer and more
suitable for end-to-end connectivity and the associated management. ADSL's rate adaption had to be described. ADSL has always been rate adaptive, but the training and management protocols used for interoperability needed standardization. The big ADSL debate over line codes has now been resolved: ETSI and the ITU have decided on DMT. Probably the only controversy to be resolved within Issue 2 is that of spectral mask, to ensure that ADSL doesn't cause any crosstalk that would prevent very high bit rate DSL (VDSL) from working.
Outside the standards bodies, there is an increasing focus on end-to-end solutions. The trend is moving toward interoperable end-to-end solutions, based on a protocol stack of PPP over ATM over ADSL. This scalable approach provides seamless connectivity, easy integration into the backbone, and a migration path to a full service network.
ADSL LITE
But as ADSL grows more stable with maturity and a growing number of shipping products, an inevitable degree of differentiation will occur. For example, one ADSL concern focuses on cost and convenience of installation: a POTS splitter, and new wiring to the computer; hence splitterless ADSL. This meshes with the idea that ADSL is "too powerful" for mass market applications Together these are described as "ADSL Lite,"an idea that has gained a lot of acceptance very quickly, Several major PC OEMs are already developing products based on it, and the ITU has a study group working on a standard, which will be interoperable with T1. The corollary is that central offices are likely to install "ADSL heavy" all of the time. Installation, training and management are significant costs, and the time scales are longer; installing a single product that supports both services will have a longer useful lifetime. Personally, I'm less convinced by the arguments in favor of low rates for three reasons: in terms of the end-to-end system the cost saving is trivial; is any amount of bandwidth really "enough?"; and will
consumers really prefer DSL at 500 kbps when cable modems promise 10 Mbps?
MORE XDSL DEVELOPMENTS
Another significant development is ADSL over integrated services digital networks (ISDN). In much of the world, ISDN is widely deployed for both data and voice services. Germany alone has installed 2.7 million lines. Many small/home of fice users want to receive ADSL data without losing their (ISDNbased) voice line. The problem is that ADSLwas planned to fit over plain old telephone service (POTS), and conflicts with ISDN signals. The proposed solution is to shift the upstream of the ADSL (above 170kHz), letting it co-exist with ISDN. Symmetric DSL (SDSL) aptly describes the (somewhat oxymoronic) Symmetric ADSL (e.g., by changing the up/down allocation of normal ADSL chips to give the same rate in each direction). Significantly, this operates over POTS and could deliver product manager 2+Mbps in each direction over the full CSA distance. The consensus regarding these developments is that we don't need to define a new set of standards, but should focus on a scalable range of interoperable options instead.

Rupert Baines is product manager for ADSL chipsets with Analog Devices (Norwood, Mass.). He can be reached at rupert.baines@ analog.com.