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 : Winstar Comm. (WCII) -- Ignore unavailable to you. Want to Upgrade?

To: SteveG who wrote (8122)	9/3/1998 1:24:00 AM
From: SteveG	Respond to of 12468

<offtopic> Both agreeing and disagreeing with different aspect of this issue, the following link courtesy of Tom Busillo (from CNBC thread): WATCHING THE MARKETS September 1, 1998 The NewsHour with Jim Lehrer Transcript pbs.org

To: SteveG who wrote (8122)	9/3/1998 1:08:00 PM
From: Bernard Levy	Respond to of 12468

Dear Steve: The 2.5bits/Hz/sec in the WCII P-MP news is more than what QPSK can offer (which is only 1.6 bits/Hz/sec). Note that it is possible to obtain signal constellations with 8 points (sometimes called PAM-PSK), so that it might be what is involved. Another fact that I would like to remind you is that it is possible to send two different orthogonally polarized (vertical and horizontal) in the same channel. This is particularly convenient to eliminate co-channel interference, if neighboring cells use a checkerboard scheme (horizontal next to a vertical cell). No one would mix compression with spectral efficiency. The 2.5bits/Hz/sec figure is based on an already compressed signal. I believe that under good weather conditions, and for moderate size cells, high order QAM (16 or 64) can be achieved at 38Ghz. This is what was posted on the Yahoo thread by a P-Com engineer working on their P-MP equipment. I was very impressed by this statement. Under adverse weather condition, you have to boost up power, but I am also aware of experiments in Japan where transmitters fall back adaptively on a lower order modulation scheme when the link deteriorates. Concerning the use of S-CDMA for the wireless ATM protocol, I have severe doubts. In checking on the literature on wireless ATM schemes, almost all schemes which have been looked at during the last 5 years are of the statistically multiplexed TDMA variety. Keep in mind also that the bandwidth available must be approximately the bandwidth for each user (here 100MHz) times the number of users. If you postulate 50 users, you would need 5GHz of BW. So CDMA is really inconvenient for broadband wireless, while it is an excellent MAC protocol for narrowband wireless. This is an area where the CDMA mantra of George Gilder is way off mark. Note also that time-slotted statistical MAC protocols are well adapted to bursty data-like traffic. So if you believe that P-MP systems will carry a lot of data traffic, the time-slot approach is preferable. Incidentally, I do not like to call these protocols TDMA protocols, since they do not allocate time slots deterministically. They are in fact in the same family as Ethernet protocols, where different users contend for resources in function of their needs. Best regards, Bernard Levy