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 : Silkroad -- Ignore unavailable to you. Want to Upgrade?

To: George Gilder who wrote (228)	2/21/1999 12:34:00 AM
From: ahhaha	Read Replies (1) \| Respond to of 626

From SR's chief technician: To add, a proprietary sampling technique combines signals into one wavelength, which is subsequently fed to an external modulator. To sample an OC-48 signal, for instance, the technique uses a frequency higher than the maximum frequency of the OC-48 signal as seen by a spectrum analyser(0-500 MHz). This creates an upper and lower sideband around the sampling frequency (the upper sideband is filtered out). To combine another OC-48 signal, the system chooses another sampling frequency with harmonics and sidebands that do not interfere with the first one. Silkroad says its sampling method does not require as high a frequency (2.2 to 2.5 times the maximum) that engineers would otherwise calculate, which enables greater information density in a given modulator's frequency domain. Using a 40 GHz modulator, Gorman says Silkroad's first system will handle 80 OC-48 signals. By aggregating the signals from multiple 980 nm lasers, he says subsequent versions of the system will scale up in capacity by orders of magnitude and still use a single wavelength. The linewidth of the beam is 175 Hz, Gorman reports, compared with 280 kHz associated with DFB lasers used for DWDM. The narrow linewidth enables longer-distance transmission without suffering from nonlinear effects compared with DWDM transmission, he says. New look at optical networking "It's pretty different from using multiple lasers in a WDM environment," Gorman says of SilkRoad's technology, noting that dispersion compensation has become a cottage industry to enable high bit rate DWDM transmission on different varieties of fiber. "The problem is with WDM in the first place," he says. "If you're not dealing with multiple wavelengths, you don't have a problem." The SilkRoad platform uses a beamsplitter to add and drop signals. "We don't have to screw around with 'lab rat' experiments with optical crossconnects," Gorman says, noting that optical switching is fast, but prohibitively expensive at $15 million for a four-wavelength lab device. To answer your question literally, no, no "unmodulation" is necessary; you split off the signal you want by its spectral output id. There is a limit to the number of inputs before saturation is reached, but until then, a signal can be added as a new sheaf in the spacetime stack. The second part of your question is the heart of the matter. It is SR's "encapsulation" of electrical to optical converter, diode detection, and laser beat frequency mixer, which accomplishes enough of a modulated boost of the signals stack that the stack will reach the destination. The detection is done almost simultaneously with the other two functions. I'm assuming you're asking about sending. On the receiving end you just have a reamplification or in the case of new input signals, if the stack isn't full, you just encapsulate add. If there is beam splitter drop of a signal, the stack amplitude drops, but that doesn't compromise the pulse reach. I think it is the case that adding and dropping cause little dispersion in the remaining signal sheafs. The sheafs are disjoint up to linewidth. How they are kept disjoint so harmonic mixing losses are avoided is only divulged to the extent of Fred Dawson's choosing. In various parts of this thread we have discussed most of the comments made here.