Texas boffins put radio waves in a spin

University of Texas researchers have developed an electronic “radio wave circulator” they say would enable full-duplex communications on mobile phones using a single frequency band. A circulator is a special application of a waveguide (using a magnetic material) that allows an RF signal to pass from one port to the next. They' …

1. At this time in the morning...

... I might need a diagram or two:

The university explains that circulators work by breaking the symmetry in a wave transmission between two points.

Cheers!

1. "Texas boffins put radio waves in a spin"

Considering the recent news about the BS claims of infinite bandwidth from infinite spin polarized RF, the headline writer has added confusion to this news item, that is on a completely separate topic.

As with planets where orbits and rotations are distinct, RF circulation is not "spin".

2. Waveguide?

Ummm... down a waveguide?

A waveguide is essentially a pipe for RF to "flow" down.

A cellphone that has to be connected back to a tower by a pipe is going to look strange.

Perhaps Apple will be able to turn it into a "must have" cool feature for fanbois.

3. This post has been deleted by its author

1. Re: allow full-duplex operation of a phone on a single frequency

Circulators have nothing to do with circular polarization. They are three-terminal A-B-C devices where signals can travel from A->B and B->C but not C->A or A->C. You connect the transmitter to A, the aerial to B, and the receiver to C. Inbound signals go from Aerial to Receiver, outbound from Transmitter to Aerial, but the high power from the transmitter never gets to the receiver (where it would wipeout any simultaneous received signal and/or damage the receiver). That way the aerial can be used for transmit & receive at the same time.

1. Re: allow full-duplex operation of a phone on a single frequency

Phil got it almost exactly correct. Some clarifications...

"That way the aerial can be used for transmit & receive at the same time...." ...on the same frequency. Antennas are used for simultaneously transmit and receive all the time, but a circulator keeps signals organized without using frequency separation.

Of course it is all nonsense. The transmitted signal is not going to find a perfect match on the antenna, especially when the user keeps moving their hands around near it. So some transmitted RF will be reflected back into the receiver. So you'd also need an LNA running with something like xxx dB of linearity. And active cancellation.

Also, making it work at a given frequency is easier than making it work over a 5 MHz bandwidth and over a half-dozen different bands.

Regarding the exaggerated and premature claims of the inventors. Keep in mind Jeffy's Rule of Tech: any bald faced lie about tech eventually becomes true. Once true, it would be revealed just how much of a bald faced lie it really was at the time.

1. Re: allow full-duplex operation of a phone on a single frequency

Well from the state that brought us Texas Instruments, who came close to beating Intel to world wide CPU dominance, I have confidence they'll work it out.

2. Re: allow full-duplex operation of a phone on a single frequency

Note that angular momentum modes can be distinguished from each other by their spatial structure, even if they have the same frequency. They are not "circular polarization modes", ie ones based on photon spin.

However, you have to line them up properly so they hit the receiver in a way that enables the spatial structure to be distinguished; i.e. more or less dead centre. This is why they use a waveguide (I assume), and also why it makes them tricky to use in a mobile device (unless you have some amazing phase-array transmitter which not only knows where the receiver is, but can dynamically tweak its output so it the signal is always aimed just right ... which would seem quite a challenging task).

However, it's worth noting that the wave circulators reported here are not using photon AM or spin; they are just isolating incoming signals from outgoing ones, but with a different technology to usual --

https://www.princeton.edu/~achaney/tmve/wiki100k/docs/Circulator.html

Nice to see a nominal breadboard still made of wood with things screwed to it. Fifty years ago we used 8BA nuts and bolts to connect the wires together - as a soldering iron's bit had to be heated in the coal fire.

Yes, it is heartwarming. Even better, if you look closely you'll see that it's an old bit of wood from the 'recycle' bin, with drilled holes from a previous use.

(Coal fire? You were lucky!)

AC " Fifty years ago ... as a soldering iron's bit had to be heated in the coal fire."

"Fifty"? 2014-50=1964? Coal powered soldering irons?

I think that you might have spelled " seventy-five" or "eighty" incorrectly.

"I think that you might have spelled " seventy-five" or "eighty" incorrectly."

I built my first transistor radio when I was 13 in 1961. It was a kit where the wires were connected together with 8ba nuts and bolts. My father had bought it as an affordable transistor radio - but it didn't work. I proved the advert's claim "so simple a child can build it".

If a home had a soldering iron - it was usually a lump of copper on the end of an iron rod with a wooden handle. It was heated either in the coal fire or over a gas hob. They were probably used for home repair to saucepans etc. Most radio valve work used fairly large components - so you could solder those with it too. There was also some "wonder" solder that you wrapped round a wire joint and heated with a burning match. My transistor experiments were usually a rats' nest of wires twisted together.

Can't remember when I was rich enough to buy my 25 watt Henley Solon** - possibly 1964/5. Until I started work at 19 my pocket money was still 1/- a week - supplemented by radio repairs for neighbours. Working with mosfet transistors with no gate protection was when I graduated to a 12 volt Antex in about 1968. By then I was earning far more money per week than my father.

Probably the reason the "Four Yorkshire Men" sketch is so popular with the older generation is that it does express a reality of the austerity 1950s childhood for many people. Grammar School emphasised the social living differences - and the fledgeling computer industry too. I remember a slightly younger colleague in the late 1970s being shocked that our house didn't have a bathroom, or even an indoor toilet, until 1970.

**Amazing that the name sprang to mind after all these years.

5. Historically the magnetic bits have been *very* hard to beat.

This is pretty impressive.

Thumbs up.

6. There's been a circulator on a project I've been involved in but I've been too busy sorting out materials to find out what the sparkies were babbling about.

"'Circulator.' I should Google that before I start specifying finishes on it..."

"Dude, drop everything and look at this drawing. One of the grognards specified cadmium plating for the fasteners with a hex chrome treatment! Erin Brockovich found out and has a mob of lawyers at the door! ...why are you reading Wikipedia at a time like this?"

So, I got edumucated today between the article and comments. Thanks, folks.

1. Spelling: 'edumakated'

7. Its a clever invention to be sure

But not going to shake up the world of mobile comms. This is always going to be dominated by the fundamental constraints of handset battery life, radio frequency allocation, and basestation costs.

8. Same frequency BAND

Your article is misleading, you cant have full duplex on the same frequency, its within the same frequency band, if you would like to re-read the article.

1. Re: Same frequency BAND

I think they are in fact claiming in their ideal world you can have full duplex on the same frequency meaning transmitting and receiving simultaneously in the same bandwidth/channel. Apart from needing a circulator to give > 100dB of isolation, and a similar level return loss from the antenna, the reciever would need to cancel the massive transmit signal reflected from the environment or users hand/head in the case of a mobile. Not to mention unrealizable linearity in everything as indicateded by others earlier.

So there are a few challenges ahead to put it mildly.

But still, I used to do stuff like this years ago and I for one, am simply awestruck at what todays engineers have accomplished, its amazing.

9. I'm getting old

I saw that picture, noticed the BNC connectors, and thought:

"That's REALLY small for a Token Ring network"

10. token ring

One of the older parts still kept here. Just in case.

11. Single frequency duplex radios and repeaters have existed for many years now, of course they do use circulators.

The circulator is just one of several tricks that have to be added together to get enough isolation, but the circulators isolation happens in the high power part, meaning that the rest of the isolation tricks only have to deal with mW or sub volt levels of the transmit signal in the receive path. This is much easier to deal with than 10W or tens of volts at the antenna path.

It is easy to imagine some uses for this, even with quite limited isolation.

- In a cell system, you could use the rx channel for very low power transmissions, that just go to the users who are nearby only. You don't need full power for all users, and this might give you 20% extra capacity

- You could use it for chip-chip comms and wired networking, halving the number of wires

- For frequency duplex, you can use two channels that are close in frequency, thus channel filtering can be done by the IF filters, not by (hughe) signal frequency duplexors. This would make duplex UHF CB's, duplex data radios, and a whole bunch of other things where you only have a single band (only cellphones get given a bunch of separated special duplex bands by the government)

- low cost, miniature radars, distance measurement systems and so on. Cellphome radar camera anyone?

12. You don't need to TX/RX on the same frequency at the same time.

Apart from the issues about the antenna match causing time varying reflections most communications networks already use a form of TDMA. In the case of TDD LTE it is the same frequency.

Same frequency with duplex TX/RX will certainly have huge multipath problems from nearby objects too so it is a nonstarter.

Interesting bit of engineering but limited in use. I'd love to see the intermodulation performance figure, bet it isn't too good.

Andy

POST COMMENT House rules

Not a member of The Register? Create a new account here.

• Add an icon

Anonymous cowards cannot choose their icon

Biting the hand that feeds IT © 1998–2019