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 : Terayon - S CDMA player (TERN) -- Ignore unavailable to you. Want to Upgrade?

To: Dan B. who wrote (226)	1/18/2000 6:18:00 PM
From: Bernard Levy	Read Replies (4) \| Respond to of 1658

Hi Dan: All, right, let's start from the basics for the anti-TERN argument. The objective of the Hi-PHY standard is to increase upstream (and if possible downstream) data rates. Increasing data rates can only be accomplished by using modulation schemes which are more spectrum efficient (high-order QAM). However, S-CDMA offers no advantage in the area of spectrum efficiency. Specifically, orthogonal CDMA (such as S-CDMA) is robust against narrowband frequency interferers, and impulsive noise. However, CDMA has nothing to offer in the area of spectral efficiency. QCOM's IS-95 spreads a BPSK signal (reapeated on both quadrature components), so that its spectral efficiency is 1bit/Hz/sec. I believe (but am not sure) that cdma2000 spreads QPSK, so that it has a spectrum efficiency of 2bits/Hz/sec (these are coarse estimates). For its Hi-PHY, I suspect that TERN was assuming that they could spread high-order QAM signals. However, a 2-year old report I saw which evaluated TERN's attempt at spreading 16-QAM fell way short of the expected 4 bits/sec/Hz efficiency. Basically, it is extremely difficult to combine the complexity of QAM transceivers with the complexity associated to S-CDMA. So, S-CDMA can at best offer marginal improvements in upstream and downstream rates. An alternative brute force approach consists of course in subdividing local user clusters as they get saturated (expensive but certainly effective). I am also aware of a solution proposed by CMTO 2 years ago (I have no idea as to its effectiveness) which essentially attempts to inroduce ``switching' in the shared cable environment. Now, let's look at it from the point of view of cable companies. Suppose you are COX, or Time-Warner, or Media One, and spent billions of dollars upgrading you cable plant to 2-way HFC, and installing DOCSIS compliant equipment. Are you going to immediately embrace a new standard that offers only marginal improvement, particularly when brute force capacity increasing techniques (such as the cluster splitting method) are available? Probably not. Now, suppose you are an operator who has not performed the 2-way HFC upgrade. Certainly TERN"s equipment is attractive. However, looking down the road, is this going to stop you from upgrading your plant at some point? No, since 2-way HFC makes possible selling a wide range of services to consumers. I never understood why the S-DMT proposal was not accepted for Hi-PHY. S-DMT has all the noise robustness features of S-CDMA, but it also comes much closer to delivering the maximum data rate sustainable by the channel. The one drawback is that it is very complex and a power hog. As I indicated in earlier posts TERN will have 100% of a market segment which ultimately will shrink to zero. Its growth rate indicates TERN is moving aggressively in filling in its market segment, but the long run prospects are quite poor. Best regards, Bernard Levy