Wifi at 10,000 Times Lower Power

if I understand the article, the savings is due to an additional layer of technology between the router and actual handheld device by using RF passive technology?

Is there power actually saved, or is it just diverted elsewhere.

I would say it is power efficient as in each device in the IoT does not need a large battery to communicate but instead piggy back in the one device plugged in the wall.

That's how I understand, the power is moved to where power is more available.

Which makes sense. I rather that, that put a drain on my batteries.

Here is the way I understand it:

They move the analog part of the processing to the router which is plugged in. The IoT devices themselves apparently send data by just switching the impedance of an antenna which modulates the reflectivity.

The claimed data rate is 11 MB/sec.

It's the same principle as communicating with flashing lights. You could have a light source at both ends or a light source at one end and toggle a mirror at the other.

Clever technique. Reminds me of a very different application: sending data back up through a drill string by modulating the flow rate/pressure of the drilling mud. They used a heavy, six-vaned tungsten carbide shutter near the bit end that would alternately cover and expose corresponding holes through which the drilling mud flowed. Very low data rates of course both because of mechanical limitations of the shutter but also because the mud column is extremely noisy in terms of pressure fluctuations. Terrible SNR which required advanced techniques to extract the data from the noise (something having to do with Bayesian statistics as I recall).

The data consisted mostly of the string electronic's basic vital signs, the bulk of the measurement data (rock resistivity, 3-axis magnetic field, neutron backscatter data to measure rock density, etc.) being saved on-board in RAM (that has to reliably withstand 200C temps and extreme shock and vibration) They measure and store a boatload of data in those strings and then download it to the rig's computers when they pull the string out of the hole.

Hey - that's encroaching on the wireline industry. What next? Perforate as they run casing?

Got it...

Hey, they should come here and test their wares. My Internet provider has offered this service for years!

I sometimes wonder if I'm paying more for the promise of Internet connectivity here than the actual fact of it. Was downloading the latest Mint Linux distro for instance and the initial ETA for the 1.2 GiB file was 4 years, 47 weeks. Be still my heart.

Very impressive that they can achieve 11 Mbps with a few microwatts. What's the error rate I wonder? Usually they have to reduce bandwidth at lower power to minimise errors. On the flip side connected devices not requiring all that bandwidth could operate at greater range.

So in reality....

What we will get is new devices that use 1/10,000th the power but they will put a battery in them that's 1/10,000th the size so we will gain nothing for overall run time.

Sounds way too plausible to me to not happen.

Some devices contain no batteries at all - nor need them - deriving their power instead from the signal itself. Many RFID devices do this, for example.

deriving their power instead from the signal itself.

that would be the term passive RF comes in

Exactly (although some readers - none here I'm sure - might read 'passive' as their being especially apathetic. "Oh gawd, not another signal." ).

Or they might have a solar cell and rechargeable battery like solar powered watches.

Cold fusion.