The relation that George S. referred to is the following:
C = B * log2( 1 + S/N )
C is channel capacity in b/s
B is channel bandwidth in Hz [1/s]
S/N signal to noise ratio
log2 is radix 2 logarithm
A flat (white) noise power spectral density and signal spectrum
is assumed.
Multiply this by a time interval T. CT becomes total bits BT is time bandwidth product which we can generalize to total number of degrees of freedom that can be independently modulated. Since this theorem is elementary probability and thermodynamics, noone can beat it including Silkroad.
George & I were trying to have an argument about what the number of degrees of freedom would be for SilkRoads device. He claims (I agreed) that as all input signals are electrically combined prior to modulation of the beam, that only the degrees of freedom associated with the frequency bandwidth of the modulator are involved. I was trying to hypothesize that additional spatial (or spacetime) modes could be involved and this was why SR's "whitepaper" discusses photon angular momentum, Laguerre-Gauss modes etc. I was doing so on the basis of SR's opaque claims coupled with the conviction that if obtaining a clean single frequency beam prior to simple AM modulation was the entirety of what SR has accomplished, George was fully justified in claiming that there is nothing here spectacular enough to get the audiences attention.
As far as I can tell the patent provides info relating only to the laser stabilization scheme developed by Palmer. Both George and I are willing, for sake of our discussion to take single mode constant frequency operation for granted. The primary feedback loop is nonproblematic as the ratio of the adjacent lines should, under suitable circumstances provide a basis for a feedback signal to keep the gain profile stabilized around the desired operating point. The extra line can be eliminated by the polarizing filter in the output coupler.
I confess to having no idea whatever regarding the outer feedback loop involving the optical phase locked loop with the "paradigm' laser. As the paradigm laser "may be another 1300 nm laser diode stabilized by thermal feedback loop similar to the transmitting laser etc. etc." Since this laser should be subject to the same first feedback loop limits as the transmitting laser it is unclear why phase locking these together helps anything. If the "paradigm" laser were to be inherently more stable, this would be another matter. Perhaps Palmer is using the paradigm laser at low power and finds that the first order feedback loop is adequate to yield high stability at low output. Since at low power diodes usually lase with a large number of modes and this input uses a 300 angstrom filter ahead of the monochromator which is roughly adequate to eliminate all but one mode, and if the this laser is also a DFB structure whose temperature dependence is primarily related to index changes then we can probably assume it to be more stable than the transmitting laser operating at much higher output.
Apart from the extraneous generation of "synchronizing pulses" for the modulator (obviously with TV signals in mind) the mechanical beam chopper appears to be simply a rather elaborate optical phase comparator for the transmitter and reference beams. The accuracy of this device is at any rate limited by the spectral resolution of the input monochromators, as discussed at length in the text.
I find nothing in the patent that addressed the modulator apart from an obscure detached portion of figure 1 which shows the modulator looking like a "thick" 45 degree reflector beam combiner, which it clearly isn't. The text does not explain why the clock carrier is "opto-electronic" rather than just plain old electronic and does not disclose the format of the "modulated arriving signal". Since none of this is discussed to any relevant extent in the text I am not clear on how the patent has furthered our understanding of Palmers arrangement as a *communications* system as opposed to a stabilized laser, nor do I find anything to adjudicate the issues being discussed between me and George.
So finaally to George ... I'll try to get back to the modulation
and elipsometric polarization content later when I have time, we
are not going to get it out of the patent.
Carl |
