Hello Jack,
Before I get to the soliton part of your question, I'd like to clear up an earlier statement I made concerning free space systems which you have resurrected here:
"1) do you have any opinion on the use of solitons as a last mile solution? they travel at the speed of light and are not as prone to dispersion and loss of signal as photons - limitations that you mentioned re lu's OpticAir system of lasers.
You have a good memory, re what I could have phrased better about the free space infrared systems. It wasn't quite accurate for me to use the term "dispersion" there without clarifying what I meant, as much as it was an expedient.
Individual pulse spreading and the spreading of the beam, itself, do take place but not for quite the same reasons through the air as through fiber, but similar. There are atmospheric influences at play which affect delivery through free space (the air, and not to be confused with non-atmospheric applications in a vacuum), namely diffusion through its gaseous components, the impacts caused by scintillation and humidity, subtle beam movements due to the instability of terminals which might be mounted on poles or on the sides of buildings which are subjected to vibrations. The latter have the effect of moving the distant "spot" or "dot" at the remote target.
One other gotcha that planners must take into account when designing "window shots" is the nature of the glass that exists in those windows. Some are very high in metallic content which can be a show stopper where distances would otherwise only be marginal, but not a problem through unobstructed free space.
All of these and other influences cause what might be viewed in terms similar to multi-path dispersion effects within a narrow field, and overall insertion loss which results in signal attenuation.
Oddly enough, the LU model you referenced actually uses a controlled form of dispersion to its advantage, and then corrects for it at the other end. The following passage is from a ZD synopsis of the OpticAir system from LU:
"This system, originally designed for a government application, transmits eight 2.5-gigabit-per-second signals through the air from a transmitter to a receiver. The system is made environmentally safe by a process that disperses, or spreads, the laser signal and uses error correction to compensate for whatever problems the dispersion causes to the information being transmitted, said Kathy Szelag, vice president of marketing for Lucent's Optical Group."
-------
[re Solitons] "...are not as prone to dispersion and loss of signal as photons.."
Photons, to the extent that we can reach agreement on just what they are, are also used in soliton technology. Solitons merely define the means by which optical pulses are encoded, as differentiated from conventional light pulsing techniques onto glass that have been used to date.
-------
"..some say solitons are not as reliable as photons and they are (presently) considerably more expensive. they'll be great for undersea and transcontinental LD networks, but are they all that much of a plus for nt's qtera, whose 2400 miles w/o regeneration technology depends on solitons?"
You've touched on the major impetuses for researching soliton technologies already. And those are its ability to overcome distances, so its use in the near term would not be for last mile applications where distance is not an issue.
Perhaps, later on, the benefits derived from a fuller understanding of solitons might enable last mile implementations through alternative materials such as POF, or plastic optical fiber, whose losses and other properties right now make them ineffective in all but the very shortest distance situations.
But by and large, solitons will be used in the short to intermediate terms to overcome the need for expensive and architecturally cumbersome optical amplification and regeneration on very long routes, especially those which are not easily reached for maintenance and modification purposes, as is the case in transoceanic crossings.
" 2) if copper has the speed limitations, coaxial the upstream limitations, and wireless or laser the reliability limitations: would the last mile solution have to be that the cable or phone companies... will finally have to connect a few strands of fibre from the backbone, and pull 'em along the poles and conduits, up through the buildings and into the livingroom? costly, but dependable, and it's just a re-run of what the cable co's just did a few years ago.
Any new pull is costly, be it fiber or twisted pair, or coaxial. The costs of fiber have historically been viewed as high due to both higher costs per element (strand) and the much higher costs which used to be associated with splicing and fiber handling, in general. But now the costs of both fiber and connectors have come down considerably, and continue to (with plateaus which are sometimes hit that last a while), but more importantly best installation practices now include fast-connects which have driven the costs of connectorization and "splicing" down, not quite into the noise area, yet, but at least competitive when viewed against other media forms today, and a hands down winner when it comes to price-performance when exploiting the next window(s) of bandwidth potential, which neither of the competing options can even come close to.
In short, the impediments to last mile fiber installation to the home are no longer the costs associated with installation (where in fact the cost per gigabit cannot be matched by any other known alternative at this point), as much as they are issues concerning forward looking compatibility with backwards looking terminal gear I/O requirements, and certain mindsets which are harder to crack than the atom was, fifty four years ago, at about the time that coaxial systems were first introduced.
HTH, and Regards, Frank Coluccio |
