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 : The *NEW* Frank Coluccio Technology Forum -- Ignore unavailable to you. Want to Upgrade?

To: Frank A. Coluccio who wrote (14701)	4/20/2006 11:36:25 PM
From: Peter Ecclesine	Read Replies (2) \| Respond to of 46821

Hi Frank, (extracts from a posting by David P. Reed on the Cook Report list) David P. Reed: Some comments on BPL "interference". First an introduction: Often my comment on "interference" gets misquoted. A precise version of my comment is as follows: "interference is not a property of radio waves, it is a property of a particular radio architecture" Without delving deeply into this, what is commonly called "radio interference" is the situation where a particular receiving device cannot distinguish among multiple independently transmitted signals. Those signals do not "interfere" with each other while propagating through the air - photons do not interact, nor do waves - but instead create a situation of potential ambiguity for the portion of the system desiring to receive one or more of them. This would not be anything more than a semantic issue, save that many engineers (including some on this list) think that what Shannon and other information theorists said is that there is a "limit" on the amount of information that can be carried in a particular region of space on a particular "frequency". That conclusion is just plain wrong - I understand where the logical error is, but to many engineers, having heard the words "Shannon" and "limit" in the same imprecise comment about "the spectrum" too many times, it somehow sounds like I am contradicting God's Book, when all I am saying is that God never said that. . . . . In truth, the fundamental issue is not energy at all, but the entropy of the system as a whole. We use energy as a proxy for entropy in our regulatory structure. This is a "conservative" approach, because one can plausibly argue that if you design your system to accept the output of a random gaussian process with energy < the energy input to that process, you have bounded the risk of a communications error. (this is NOT the most conservative bound - for those who understand the "byzantine generals" game, but it's pretty conservative). The key to understanding how radio networks can scale involves designing to manage the interaction among intentional signals, which have a controllable amount of entropy or randomness, from the point of view of receivers that receive them. If you can't do that, they you have to assume signals you are listening are combined with other signals which can be anything. We "conservatively" model "anything" by a gaussian random process with an energy bound. . . . Then more details about BPL, but I think this part is more than enough to illuminate some of the difficulties in regulating 'interference' (Section 302 of the US Communications Act of 1934). In another context, wirelessUSB must coexist with Wi-Fi on the same laptop, so the wirelessUSB receivers have to be prepared for ~120 dB of dynamic range of in-band energy, or they will suffer the same fate as Bluetooth 1.0 receivers petere

To: Frank A. Coluccio who wrote (14701)	4/22/2006 3:20:42 AM
From: axial	Read Replies (2) \| Respond to of 46821

Frank, ftth, I'm not sure I can make a network-savvy reply here. It seems to me that common sense and simple observation suggest some realities: 1 - Any finite transmission medium can eventually become saturated with traffic. 2 - While the medium is finite, potential traffic may not be. 3 - At the saturation point, experience suggests that we resort to a heiracharchy of remedies: A - Differentiation and prioritization of traffic using any of several metrics B - Increased capacity C - Alternative transport If you'll excuse the comparison, our highways offer a good example of undifferentiated traffic. The constraints on additional capacity are cost, and a host of other considerations, which boil down to whether the additional capacity is advisable. So we institute tolls, special lanes for high-occupancy vehicles, and so on. Or we improve capacity. Or we put people on trains and subways. Interesting that some transmission media do not permit "A" - a pipeline carrying treated water, gas, or oil can only be remedied with "B" or "C". That's true because differentiation has already been achieved. I'm just as quick as others to revile capacity constraints dictated by the profit line, or some other questionable criterion. But it seems to me that differentiation and prioritization are an inescapable reality of network usage, and User Pay will be the most "convenient" means of delaying or avoiding capital expenditures for capacity increases. Jim