It was cool until the 2.5cm to theoretical 5m range was mentioned.
Shrink the process node, and maybe the power requirements will go down and the range can be extended again.
Dutch researchers have invented an internet-of-things sensor that powers itself from router radio waves. The first sample chip measures just 2mm square, weighs 1.6 milligrams and measures temperature. It draws power from a Wi-Fi router via a tiny antenna, takes a reading, and then broadcasts it back, using a slightly different …
Quite.
Lack of range makes the suggested use of "Building-wide temperature maps" a complete non-starter.
Even at the theorised 5m range, if you need a router every 5m to supply enough power to the sensor, obviously you need to get power to all those routers.
The chips might cost 20c each, but sufficient routers, their power cabling, ducting etc certainly costs more.
Cheaper and easier to just use wired sensors. Less wifi noise too. Imagine the channel interference with a router every 5m...
Neat that it can power itself. Pretty much useless in situations where you want blanket coverage.
Imagine a future in which a device along these lines has a remotely exploitable bug, and you can't shut the darn thing off as a last ditch mitigation because you've slapped a few thousand of these doohickeys into a building's fabric, and there's no easy way to power only the "good" IoT bugs with RF without also powering the "bad" ones.
Dave: "HAL, turn on the heat in here, it's freezing!"
HAL: "I'm sorry, Dave. I'm afraid I can't do that. The temperature sensors indicates that the internal temperature of your room is 127°C."
I wonder if their next invention will be some kind of electronic musical instrument that can be used to make spooky music for SciFi films[*].
Leon Theremin's "The Thing" is this-a-way:
https://en.wikipedia.org/wiki/Thing_%28listening_device%29 [**]
* Interestingly, Dimitri Tiomkin's soundtrack for Howard Hawks' "The Thing from Another World", uses a theremin.
** I have no idea how to embed links. So there.
Maybe the chip is 20 cents, but then there's repeaters every few metres which will bump the effective cost up to, errm, whatever a repeater costs, maybe $20, so the chip cost is an irrelevance !
Plus add the cost of installing the power cabling to run the repeaters, and the hassle of that.
So you might as well, as a comment above says, just run wires and embedded sensors.
Back to the drawing board I think.
I've never liked the whole 'harvest power from radio waves' idea since you still need to produce the power anyway and converting it to/from radio waves is ridiculously inefficient (Not to mention the interference caused by bumping the power up). Wouldn't it just be easier to use a layer of conductive paint and harvest energy by way of micro pulses? I figure that you'd place these along the top/bottom edge of the wall or in corners, so a thin strip of paint is all that would be needed. You could even base it off the 1-wire bus standard so that the conductive strip also transfers the information.
I actually really like the idea. It sucks out the background waves, so actually if you have enough of these in the walls then wifi will improve immeasurably because your neighbour will be shielded from you. This kind of thing also has the potential to create a faraday cage that powers your house - quiet and cheap :)
>This kind of thing also has the potential to create a faraday cage that powers your house - quiet and cheap :)
I assume you're being sarcastic, but this only works in the same sense that wind turbines will stop hurricanes, or solar panels will plunge the world into darkness.
In other words, it doesn't.
What it will do, though, is give the EM-sensitivity crowd a headache (another, anyway), trying to work out if it's good or bad.
Um, it is dark under solar panels, and the wind is slower on the other side of a turbine. That's how both of those actually work and this tech takes EM and turns some of it to power, removing that power from the ether. I'm not being sarcastic, I just seem to have a better grasp of physics than you do
Tesla's problem was that radio energy dissipates quadratically with distance: as in to provide the same amount of power twice as far away you'd need four times the power. In other words, for any practical distance, the amount of power you can draw has to be pretty tiny, which is what this is trying to achieve.
Presumably, like iButton and similar devices they'll all come with a unique serial number laser cut into the chip during production or set by OTP fuses which can be blown during software install and testing. The technology to do that exists today and has been around for years.
The 5m is optimistic, with RF quiet environment and large direction receiver, unless this is a perpetual motion machine breaking the laws of physics.
A 2.5cm to 20cm range is more realistic. It doesn't matter how little power the chip uses. Even if it has a BIG capacitor and only transmits in 1:100 duty cycle low data rate bursts.
You can't fit any kind of sensible WiFi aerial on a 2mm x 2mm chip either.
This is no different to an RFID chip, simply sending temperature rather than an ID Tiny range unless less specialist kit to power and talk to it and listen to it. "Smart Building" uses are fantasy.
I seem to recall that a farmer rigged up some fluorescent tubes in a barn that leeched power from, I think, the broadcast of a nearby TV tower (it might have been the leakage from power lines, any confirmation either way would be appreciated). I think that complaints of lack of signal from people in the "shadow" eventually helped the relevant powers that be locate the source.
The charge might have been abstraction of electricity but again I am not sure of what they did him for, or if he was just given words of advice. I believe I read about this incident on this might organ but cannot find it to use as a reference.
If I'm not too far wide of the mark, then might this tech, in the UK at least, fall foul of the same law? I think the issues regarding "domestic" wifi are few but what if it affects a public wifi signal or even mobiles?
In early radio much receiving gear was home made and a favourite kids project was to wrap wire round a pickle jar to build a long wave receiver that didn't need power. It could only drive audio through an earpiece though, and loudspeaker would need a powered amplifier circuit.
I think just about all the stories about people harvesting useful power from RF belong in the urban myth category.
http://www.telegraph.co.uk/technology/3336114/Over-to-you-Mythical-electricity.html
Sounds like variants on an (sub)urban myth. Anyway, it's been covered by the MythBusters. Any amount of power you could draw from being ambient radio towers or power lines will be limited by the distance and the impractical amounts of wire you'd need to leech enough energy to power a watch As Adam Savage put it, "You might as well just go get a battery."
That said, there can be situations where batteries would be impractical, such as very remote or embedded (in the literal sense) applications.
This post has been deleted by its author
I vaguely remember something from the '80s about Texas Instruments fabbing the circuit of a Tesla coil onto a chip.
The thing ended working like a broadband RF receiver and could pull a little power from radio waves.
No idea what happened to it ... (looking over my shoulder at the Union Carbide man)
I wonder if it would also be possible to use a tiny PV cell. Assuming it's embedded in paint then a small amount of light will penetrate into the paint possibly enough to power the device - it can't require very much power after all. In the future I could imagine devices that are powered by decomposing random organic molecules that land on them although that might not work so well if they are embedded in paint!