When we talking about antenna efficiency, it depends on the resistance ratio:
η = Rrad / (Rrad + Rloss)
where
Rrad - radiation resistance
Rloss - loss resistance
Rloss = Rs + Rd + Rn
where
Rs - resistance due to skin effect
Rd - resistance due to dielectric loss
Rn - resistance due to heat loss on bodies placed into Near Field region
In other words, antenna efficiency is a ratio between EMF radiation loss and heat loss.
Right. Suppose we have a capacitive plate, or wire, or top load (to use the Tesla equivalent) that has an impedance of several megohms. Suppose that it's mostly radiative loss (i.e., say 5MΩ radiation loss in parallel with 20MΩ equivalent parallel loss).
But it's such an extremely high impedance, that we have no hope of tuning it as an antenna, with an LC tuner. Right?
Right.
Suppose then, we magically have a resonator, with a Q of 300,000, with a tap having an impedance of 5MΩ. Its equivalent parallel loss at that tap is therefore 5MΩ * 300,000 = fuck all resistance. Suppose it also has a tap down around, maybe not 50Ω, that would be a huge ratio, but say it's kΩ where we can still match to it reasonably with an LC tuner. Or maybe 50Ω is reasonable after all, it's a huge Q -- I don't know. Just mechanical tweaks really.
Finally, suppose we wire these components together.
Am I to understand that, just because the resonator has
some kind of loss whatsoever, that now, all of a sudden, it will be the dominant loss in the system, even though its equivalent loss is three hundred thousand times weaker than the load we've attached to it?
I don't know about that.
Hm, may be it's true, then this antenna is just attempt to replace impedance match circuit (antenna tuner) with a better efficiency crystal. But there are still other issues exists. The crystal should be as large as possible in order to get better efficiency. This may be problematic for crystal. And such antenna still should be placed far away from the ground in order to eliminate heat loss on the bodies falls into reactive near field region. It may be useful for spacecraft or weather balloon, because they are far away from the ground.
There we go.
Yes, the crystal only needs to be a resonant transformer. It will inevitably have a high Q, which is possible thanks to its good mechanical properties and symmetrical mounting.
Ground effects can be ameliorated somewhat by using a dipole (nothing stopping one from using a symmetrical resonator as a combination balun and transformer!), but it will still depend upon nearby equivalent dielectric losses. In effect, the radiation resistance as seen by the antenna might be dominant (giving high efficiency into the fields, near and far), but the radiation resistance, as integrated over the near field volume, might still include a lot of losses within that volume, so that the far field radiation still ends up a small fraction of it.
On the upside, the impedance is very much higher than the ground impedance, which is not the case for a magnetic loop antenna, where the ground impedance is a sizable fraction. That is to say, the mag loop might have an impedance of fractional ohms, which is merely 1/1000th of the ~100s ohms ground impedance; but this antenna will have megohms, which is 10,000s (or more) times the ground impedance, so the efficiency can be that fraction higher.
Also, so high Q factor leads to low speed communication, it will be about 0.05 baud or something like that for 30 kHz.
Yes, so now we can come back to the original paper: by parametrically varying the resonator, transmission rates for a particular modulation (namely, FM, with some inevitable AM as well) can be much higher than the Bode-Fano limit of a LTI antenna.
Reception rates won't be improved, or at least, you could vary the parameter in lockstep with what you're receiving and pick it up correctly, but, how would you know what you're receiving unless you've already received it? D'oh. So, no help there. Such symmetry violations are the price of exceeding the Bode-Fano limit.
The proof of the pudding is ultimately in the eating, of course: if, say, the military isn't using this technology, it must not be worthwhile. They wouldn't need it for airborne transmitters, because they have line of sight to any band they wish (ground, aviation, satellite..). It would be useful for subs, but only when they surface (the impedance of ocean is even better, ~10s ohms, by the way), otherwise the E-field is shorted out by whatever's around it of course. And again, they can do that with a much more effective wideband antenna, if they're going to risk exposing anything above the surface at all.
If it truly is more efficient as a transmitter, and can be scaled up to considerable power levels, it may prove cheaper to operate than existing methods. That may take time to implement though, so maybe the military isn't so hasty to adopt it.
Tim