Comment by xraystyle
5 days ago
This comment really sums it up well. Literally everything with antenna design is a trade-off. You can design an antenna to radiate very well at a given wavelength. The better it is at doing this, the worse it tends to be at every other wavelength. You can make an antenna that radiates to some degree across a wide array of wavelengths, but it's not actually going to work very well across any of them.
Same thing with radiation patterns. You can make a directional antenna that has a huge amount of gain in one direction. The trade-off is that it's deaf and dumb in every other direction. (See a Yagi-Uda design, for instance.)
Physics is immutable and when it comes to antenna design there really is no such thing as free lunch. Other than coming up with some wacky shapes I don't really think AI is going to be able to create any type of "magic" antenna that's somehow a perfect isotropic radiator with a low SWR across some huge range of wavelengths.
> perfect isotropic radiator with a low SWR across some huge range of wavelengths
Fair analysis -- but of course, there are industries where a funky and expensive radiator optimized for a single frequency could be very worthwhile.
That's the thing though, is that it's not hard to make a good antenna for a single frequency. We already know exactly how to do that. And when we're talking transmission and reception of radio, tiny incremental gains that might be eked out through some wacky design generally don't move the needle very much.
I can talk to the astronauts on the ISS on 2 meters with an antenna I can make out of a PVC pipe and a metal measuring tape using a 5-watt transmitter. Improving that design by 2% doesn't really mean anything useful in this context.
It would usually be vastly cheaper and easier to just increase the transmit power. Or sometimes it's the available power that's the limiting factor, and a 2% increase to the antenna isn't going to matter.
Point is, trying to chase tiny gains in one dimension or another over a thoroughly tested and well-understood antenna design is kind of a waste of time outside of an academic, beard-scratching context.
Well, the speculation is that an AI could iterate through a zillion novel and mostly-garbage designs to discover something unexpected with a higher gain than known designs.
There's a percent efficiency/gain improvement that exceeds the cost-performance ratio of simply increasing power -- boiling down to the usual capex vs opex argument.
I can't make an intelligent guess on the likelihood of that discovery.
There's always a market for a better free lunch.