These designs fascinate people who haven't designed antennas. I don't doubt that throwing enough computational power at optimizing antennas will produce antennas optimized for something at the expense of something else but if you're a casual what you should notice is that these papers never mention the "something elses". You can get a paper out of just about any antenna design, btw. There's also a type of ham that will tune up a bedframe or whatever. So just getting something to radiate should not be confused with advancing the state of the art.
These antennas found their way into the utterly savage "pathological antennas" chapter of Hansen and Collin's _Small Antenna Handbook_. See "random segment antennas". Hansen and Collin is the book to have on your shelf if you're doing any small antennas commercially and that chapter is the chapter to go to when you're asked "why don't you just".
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.
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.
"It has become fashionable to design wire antennas with some type of optimizer program, almost independent of good physics or high-quality performance. The results sometimes have wire segments in all directions; see Figure 5.24 for an example. A long total wire length may achieve resonance in a small volume, but there are several disadvantages. If Z is the normal monopole direction, the X currents tend to cancel, as do the Y currents. However, in certain directions the cross- polarized field may not be negligible. Longer total wire length increases loss resistance, reduces efficiency, and increases reactance. And generally the bandwidth is narrow. Examples are Altshuler and Linden (2004), Choo et al. (2005), Altshuler (2005), and Best (2002, 2003). Use of fractals and meanderlines to fill space (Gonzalez-Arbesu ́ et al., 2003; Best and Morrow, 2002) suffers from the same problems.
“Do not confuse inexperience with creativity” (Linda Whittaker) is appropriate here."
Not only are they pathological, but when you order a example be built because CST confirmed that the design would kick ass and you put it in the chamber and actually measure it for real, you walk away wondering why you wasted so much time and money.
I wish I'd had a copy of the book referenced much earlier in my career too.
As far as the twisted-paperclip antennas go, just imagine trying to verify each of those 3D bends was to spec. Or conversely, running a monte carlo on all the degrees of freedom in that design.
I just read it in the referenced book section from the parent comment. It shocked the imaginary bubble where my mind is a bit. I want to reflect more on it.
Somehow, in the midset of all these LLM and diffusion models, the only thing that seems to catch attention is creativity. I've not thought of experience.
Very cool. Evolutionary Algorithms have kinda been out of the mainstream for a long time. They are good when you can do a lot of "black-box" function evaluations but kinda suck when your computational budget is limited. I wonder if coupling them with ML techniques could bring them back.
> I wonder if coupling them with ML techniques could bring them back.
EAs are effectively ML techniques. It's all a game of search.
The biggest problem I have seen with these algorithms is that they are wildly irrespective of the underlying hardware that they will inevitably run on top of. Koza, et. al., were effectively playing around in abstraction Narnia when you consider how impractical their designs were (are) to execute on hardware.
An L1-resident hill climber running on a single Zen4+ thread would absolutely smoke every single technique from the 90s combined, simply because it can explore so much more of the search space per unit time. A small tweak to this actually shows up on human timescales and so you can make meaningful iterations. Being made to wait days/weeks each time you want to see how your idea plays out will quickly curtail the space of ideas.
An L1-resident algorithm can outperform one that needs to talk to DRAM each iteration by 100x or more. In terms of wall clock time, this can mean the difference between minutes and days.
Would you be willing to try a fleeting idea if it took 2 days to test? How about if we could bring that down to 15 minutes?
The main use of evolutionary algorithms in machine learning currently is architecture search for neural networks. There's also work on pipeline design, finding the right way to string things together.
Neural networks already take a long time to train so throwing out gradient descent entirely for tuning weights doesn't scale great.
Genetic programming can solve classic control problems with a few instructions when they can solve it, so that's cool.
They're not in the press a lot. They're probably still in production behind the scenes. I was reading about using them for scheduling not long ago. Btw, a toy one I wrote to show how they work got best results with tournament selection with significant mutations (closer to 20%).
There's a lot of problems where you're searching among many possibilities in a space that has lots of pieces in each solution. If you can encode the solution and fitness, a GA can give you an answer if you play with the knows enough. You also might not need to be an expert in that domain, like writing heuristics. If you know some, they might still help.
Much more robust than almost all modern ML algorithms which let's be real, aren't exactly applicable to anything outside recommendation systems and 2D image processing.
Genetic algorithms' weaknesses largely boil down to getting stuck in local extrema and premature convergence, which can be resolved by trying different values for parameters like probability of mutation, trying different genetic operators, offspring/parent ratio etc.
Meanwhile you have a whole separate discipline [1] for potential weaknesses on machine learning algorithms. Of course they may win when it comes to interdisciplinary ubiquity in CS, but any algorithm that relies on data assimilation and has little analytic formulation will suffer in robustness.
As somebody who almost fried his computer during the antenna design course to optimize a dipoles array with a (not optimized) genetic algorithm, I really like this content.
There are many different types of antennas, each with different tradeoffs. Some examples of the tradeoffs are:
- How well does it work at a specific frequency, if you're just trying to transmit/receive on one specific frequency
- How well does it work on the frequency range(s) if you're working on more than a specific frequency
- How well does it block frequencies that you don't want to send/receive
- How well directional is it to trade off using lots of radiation to blast in many directions vs a higher focus beam using less energy or getting less interference from other directions
- How much physical space do you have in each dimension?
These are just a few examples, but for example you can provide a much "better" connection in almost every sense of the word if you can make your antenna directional (point between the source and destination) only on a specific frequency, and be huge, but most of the time you have some physical space constraints, multiple frequencies to deal with, and the potential that your signal could at least come from some degrees in each the x/y/z axes, and sometimes it needs to be omnidirectional.
Again, these are just examples, but you end up with these types of design considerations that play into larger system design (can you put more transmitters up to encourage directionality, limit frequencies, etc).
There are some well known "base" antenna types like dipole, yagi-uda, circular, and log periodic dipole array if you want to look them up by name and see some of the known tradeoffs and design choices, but virtually any wire can be an antenna and there are an unlimited number of shapes, nearly all of which don't have known radiation characteristics
The requirements and constraints depend on the product.
For example, the geometry of your product to accommodate the antenna is not always the same; the (internal and external) environment of the case is also different; there may be requirement of combining various frequency bands into one antenna; etc.
Right. Everything in the near field of the antenna is part of the antenna. Handheld products are an extreme example of this: remember "just avoid holding it in that way"? (after that, Apple built big test chambers and brought journalists in to see them)
You can claim that you aren't a highly-skilled psycho-super-soldier developed to infiltrate comment sections to spin engineers into a frenzy, but I won't believe it.
In my limited understanding, there are many factors differentiating between antennas, different antennas are better at emitting/receiving at different frequencies, and also there's directionality in the mix . For example a satellite dish and an FM radio antenna are both antennas, they're certainly not the same thing.
A long time dream of rocket scientists is single-stage-to-orbit. Ideally you'd have a vehicle that takes off and lands like a conventional jet plane at a regular airport. I've always thought that perhaps AI and evolutionary algorithms might be able to navigate a way through the various tradeoffs and design constraints that have stopped us so far.
As an avid observer of rocket design, I suppose that hasn't happened because SSTO may not have any good solutions. I further suppose that the design parameters are so constrained there is very little opportunity for a generative or evolutionary, or any other AI-driven design approach, to do more than optimize some components.
SpaceX's solution with Starship definitely demonstrates how difficult a problem it is. Raptor are the best engines humanity has, all the landing hardware is part of the launch tower to save weight, and those seem to be table stakes. Stoke aerospace has a fantastically genius solution with their regeneratively cooled heatshield / expander cycle aerospike engine. It literally turns the energy you're trying to burn off during re-entry into delta-v in the opposing direction while reducing weight and complexity.
There are a few software options available, I think, from simple little Java runtime executables to enterprise suites[1] costing 5-6 figures to license.
Could somebody share link to calculator of antenna parameters for arbitrary shape? Or may be some good book, where to read, how these things calculated?
I've seen few books on this topic, but have some issues on translate them into program.
A good rule of thumb: never mock someone’s enthusiasm or excitement about learning something, even if it’s old news to you. Let people enjoy discovering things.
I'm rediscovering it. I remember reading about this in some flashy, superlative Popular Science article, from the early/mid 2000s. So I was quite excited to click on the link and see that shape again.
These designs fascinate people who haven't designed antennas. I don't doubt that throwing enough computational power at optimizing antennas will produce antennas optimized for something at the expense of something else but if you're a casual what you should notice is that these papers never mention the "something elses". You can get a paper out of just about any antenna design, btw. There's also a type of ham that will tune up a bedframe or whatever. So just getting something to radiate should not be confused with advancing the state of the art.
These antennas found their way into the utterly savage "pathological antennas" chapter of Hansen and Collin's _Small Antenna Handbook_. See "random segment antennas". Hansen and Collin is the book to have on your shelf if you're doing any small antennas commercially and that chapter is the chapter to go to when you're asked "why don't you just".
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.
"It has become fashionable to design wire antennas with some type of optimizer program, almost independent of good physics or high-quality performance. The results sometimes have wire segments in all directions; see Figure 5.24 for an example. A long total wire length may achieve resonance in a small volume, but there are several disadvantages. If Z is the normal monopole direction, the X currents tend to cancel, as do the Y currents. However, in certain directions the cross- polarized field may not be negligible. Longer total wire length increases loss resistance, reduces efficiency, and increases reactance. And generally the bandwidth is narrow. Examples are Altshuler and Linden (2004), Choo et al. (2005), Altshuler (2005), and Best (2002, 2003). Use of fractals and meanderlines to fill space (Gonzalez-Arbesu ́ et al., 2003; Best and Morrow, 2002) suffers from the same problems.
“Do not confuse inexperience with creativity” (Linda Whittaker) is appropriate here."
Who? I can't find the source, and it seems everybody knows about it.
EDIT: Oh, it's the book itself. But what is _their_ source?
Are you asking for Collins and Hansen’s sources?
Not only are they pathological, but when you order a example be built because CST confirmed that the design would kick ass and you put it in the chamber and actually measure it for real, you walk away wondering why you wasted so much time and money.
I wish I'd had a copy of the book referenced much earlier in my career too.
As far as the twisted-paperclip antennas go, just imagine trying to verify each of those 3D bends was to spec. Or conversely, running a monte carlo on all the degrees of freedom in that design.
Antenna design feels like some occult arts
It kind of is. If your SWR is 1.09, in theory you could do better, but in practice, there's generally nowhere to go but up.
Any chance I could sell you a high-priced "cryogenically-treated" length of coax?
"Do not confuse inexperience with creativity"...
I just read it in the referenced book section from the parent comment. It shocked the imaginary bubble where my mind is a bit. I want to reflect more on it.
Somehow, in the midset of all these LLM and diffusion models, the only thing that seems to catch attention is creativity. I've not thought of experience.
Experience makes creativity harder, but that's what mature creativity is. Did anyone tell you it wouldn't be work?
The people who are most awed by LLMs are those people most used to having to be merely plausible, not correct.
Very cool. Evolutionary Algorithms have kinda been out of the mainstream for a long time. They are good when you can do a lot of "black-box" function evaluations but kinda suck when your computational budget is limited. I wonder if coupling them with ML techniques could bring them back.
> I wonder if coupling them with ML techniques could bring them back.
EAs are effectively ML techniques. It's all a game of search.
The biggest problem I have seen with these algorithms is that they are wildly irrespective of the underlying hardware that they will inevitably run on top of. Koza, et. al., were effectively playing around in abstraction Narnia when you consider how impractical their designs were (are) to execute on hardware.
An L1-resident hill climber running on a single Zen4+ thread would absolutely smoke every single technique from the 90s combined, simply because it can explore so much more of the search space per unit time. A small tweak to this actually shows up on human timescales and so you can make meaningful iterations. Being made to wait days/weeks each time you want to see how your idea plays out will quickly curtail the space of ideas.
> A small tweak to this actually shows up on human timescales and so you can make meaningful iterations.
Please could you explain what you meant by this part? I'm trying and failing to understand it.
An L1-resident algorithm can outperform one that needs to talk to DRAM each iteration by 100x or more. In terms of wall clock time, this can mean the difference between minutes and days.
Would you be willing to try a fleeting idea if it took 2 days to test? How about if we could bring that down to 15 minutes?
Ok, I think I get you now. This sentence was in context of the previous one, about an algorithm that never needs to leave L1.
I thought you were referring to some real world (I mean physical world) example of evolution finding efficiency in some manner.
The main use of evolutionary algorithms in machine learning currently is architecture search for neural networks. There's also work on pipeline design, finding the right way to string things together.
Neural networks already take a long time to train so throwing out gradient descent entirely for tuning weights doesn't scale great.
Genetic programming can solve classic control problems with a few instructions when they can solve it, so that's cool.
They're not in the press a lot. They're probably still in production behind the scenes. I was reading about using them for scheduling not long ago. Btw, a toy one I wrote to show how they work got best results with tournament selection with significant mutations (closer to 20%).
There's a lot of problems where you're searching among many possibilities in a space that has lots of pieces in each solution. If you can encode the solution and fitness, a GA can give you an answer if you play with the knows enough. You also might not need to be an expert in that domain, like writing heuristics. If you know some, they might still help.
We typically would solve a lot of the same types of problems with RL today because it’s more efficient.
In EA if a candidate fails we throw it away. In RL we learn from that experience.
RL gets harder when rewards are really sparse. OpenAI developed evolution strategies which is a bit of a hybrid.
Much more robust than almost all modern ML algorithms which let's be real, aren't exactly applicable to anything outside recommendation systems and 2D image processing.
I can't tell if this is a joke
Genetic algorithms' weaknesses largely boil down to getting stuck in local extrema and premature convergence, which can be resolved by trying different values for parameters like probability of mutation, trying different genetic operators, offspring/parent ratio etc.
Meanwhile you have a whole separate discipline [1] for potential weaknesses on machine learning algorithms. Of course they may win when it comes to interdisciplinary ubiquity in CS, but any algorithm that relies on data assimilation and has little analytic formulation will suffer in robustness.
[1] https://en.wikipedia.org/wiki/Adversarial_machine_learning
There is no reason I couldn’t use the same adversarial attacks against an EA. It just doesn’t have a Wikipedia page because EA isn’t as common.
No. The point is that the attack surface is more vast for data-driven models.
As somebody who almost fried his computer during the antenna design course to optimize a dipoles array with a (not optimized) genetic algorithm, I really like this content.
Can someone share with me why antennas need to be designed in the first place? I thought it was one-shape-rules-all type of problem
There are many different types of antennas, each with different tradeoffs. Some examples of the tradeoffs are:
- How well does it work at a specific frequency, if you're just trying to transmit/receive on one specific frequency
- How well does it work on the frequency range(s) if you're working on more than a specific frequency
- How well does it block frequencies that you don't want to send/receive
- How well directional is it to trade off using lots of radiation to blast in many directions vs a higher focus beam using less energy or getting less interference from other directions
- How much physical space do you have in each dimension?
These are just a few examples, but for example you can provide a much "better" connection in almost every sense of the word if you can make your antenna directional (point between the source and destination) only on a specific frequency, and be huge, but most of the time you have some physical space constraints, multiple frequencies to deal with, and the potential that your signal could at least come from some degrees in each the x/y/z axes, and sometimes it needs to be omnidirectional.
Again, these are just examples, but you end up with these types of design considerations that play into larger system design (can you put more transmitters up to encourage directionality, limit frequencies, etc).
There are some well known "base" antenna types like dipole, yagi-uda, circular, and log periodic dipole array if you want to look them up by name and see some of the known tradeoffs and design choices, but virtually any wire can be an antenna and there are an unlimited number of shapes, nearly all of which don't have known radiation characteristics
The requirements and constraints depend on the product. For example, the geometry of your product to accommodate the antenna is not always the same; the (internal and external) environment of the case is also different; there may be requirement of combining various frequency bands into one antenna; etc.
Right. Everything in the near field of the antenna is part of the antenna. Handheld products are an extreme example of this: remember "just avoid holding it in that way"? (after that, Apple built big test chambers and brought journalists in to see them)
Where are the antennas on your phone today?
You can claim that you aren't a highly-skilled psycho-super-soldier developed to infiltrate comment sections to spin engineers into a frenzy, but I won't believe it.
In my limited understanding, there are many factors differentiating between antennas, different antennas are better at emitting/receiving at different frequencies, and also there's directionality in the mix . For example a satellite dish and an FM radio antenna are both antennas, they're certainly not the same thing.
A long time dream of rocket scientists is single-stage-to-orbit. Ideally you'd have a vehicle that takes off and lands like a conventional jet plane at a regular airport. I've always thought that perhaps AI and evolutionary algorithms might be able to navigate a way through the various tradeoffs and design constraints that have stopped us so far.
As an avid observer of rocket design, I suppose that hasn't happened because SSTO may not have any good solutions. I further suppose that the design parameters are so constrained there is very little opportunity for a generative or evolutionary, or any other AI-driven design approach, to do more than optimize some components.
SpaceX's solution with Starship definitely demonstrates how difficult a problem it is. Raptor are the best engines humanity has, all the landing hardware is part of the launch tower to save weight, and those seem to be table stakes. Stoke aerospace has a fantastically genius solution with their regeneratively cooled heatshield / expander cycle aerospike engine. It literally turns the energy you're trying to burn off during re-entry into delta-v in the opposing direction while reducing weight and complexity.
As a rocket scientist I assure you it's been tried
This has to have been done in more modern times in simulation of the EM field for a better design instead of practically
There are a few software options available, I think, from simple little Java runtime executables to enterprise suites[1] costing 5-6 figures to license.
1. https://www.keysight.com/us/en/products/software/pathwave-de...
Could somebody share link to calculator of antenna parameters for arbitrary shape? Or may be some good book, where to read, how these things calculated?
I've seen few books on this topic, but have some issues on translate them into program.
Do people not go on Wikipedia nowadays? This is literally on the frontpage of the wiki for this stuff: https://en.wikipedia.org/wiki/Genetic_algorithm
A good rule of thumb: never mock someone’s enthusiasm or excitement about learning something, even if it’s old news to you. Let people enjoy discovering things.
I'm rediscovering it. I remember reading about this in some flashy, superlative Popular Science article, from the early/mid 2000s. So I was quite excited to click on the link and see that shape again.
But also, something something lucky ten thousand.
https://xkcd.com/1053/
I can't help myself, I always flip to the last page of a Wikipedia article to find out who did it.