Back in the days of 2G the pre-wake-up pulses, aside from causing massive FM/AM interference like demented Morse code, would light up tuned antenna spark gaps on stickers for your Nokia phone. In Japan they sold cute phone tokens which had glowing eyes.

Energy harvesting the same way Theremin did for his passive wall bug.

Reminds me of an old Nokia proof of concept charger for a mobile phone that used energy harvesting. It never was enough to make a call though:

From 2009 https://www.eetimes.com/nokia-working-on-energy-harvesting-h...

But it seems that the idea is still alive, from 2023:

"Relying more on ambient energy sources could prove monumental in automated warehouse inventory tracking, in medical instrument management and for deployment in airports, shopping centers and even individual smart homes. Nokia’s goal is to have energy harvesting technology in cellular networks that can support this massive IoT deployment."

https://www.nokia.com/blog/the-future-belongs-to-zero-energy...

I read the technical description of this and the whole work seems like magic. Antenna design, extremely low-power passive networks, etc. On top of that, it can tap into network signal bursts as a communication medium.

It feels like a project sent from the future.

I've always been curious what energy harvesting systems are capable of.

Also, what is the third type of energy harvesting besides light and 2.4GHz? I couldn't figure out what that might be.

For a while I had a very overpriced phone case from moeco that lit up when cell service was being used - https://www.moeco.jp.net/ - it came with a big warning not to use wireless charging with it (due to instead heating the case). Sadly, I wouldn't get one nowadays, I love wireless charging too much.

This is super neat. By modulating the timing of the data being sent to a websocket, (which is basically a /dev/null data sink) it implements a covert air-gaped side-channel data transmission mechanism.

> That last one consumed an hour of diagnostic time and involved using time-domain reflectometry (with a 20 ps rise-time pulser and 20 GHz scope) to locate the fault to within a region of a couple millimeters on one trace.

How does one even obtain the skills, much less the equipment to run such precision?!

Interesting, thanks for sharing.

One question: what load is the matching network designed for? Did the designer find the equivalent small signal impedance of the diode network via simulation? Is the SPICE model even valid at 2.4 GHz? Is small signal even applicable?

This is the same principle as those mobile phone led flashing stickers from the 90’s/00’s.

Clever idea.

very nice! They mention the use of Barker codes, which I'd not heard of before;

https://en.wikipedia.org/wiki/Barker_code

This project is so inspiring. I just spent about a half hour+ oogling over the details. Great documentation and I learned about so many things.

Very cool magic/tech!

Apologies if this goes against the hacker spirit, but do you know where I can buy a similar thing? In Europe?