Passive WiFi: The Fine Print

News about the passive WiFi breakthrough circled and confused the world. It is supposed to use 10 thousand times less power compared to standard WiFi. Yet still being 100% compatible, as demonstrated on the video. People keep repeating the story on and on and not many took the effort to read the original paper and especially the fine print.

So what is in the fine print?

Here are the quotes:

"...the power-consuming RF components such as frequency synthesizers and power amplifiers are delegated to a single plugged-in device in the network. This device provides the RF functions for all the passive Wi-Fi devices in the vicinity by transmitting a single-frequency tone..."
"...we set the plugged-in device to transmit its tone at a frequency that lies outside the desired Wi- Fi channel..."
"We set Pt , Gt , Gr and Gpassive to 30 dBm, 6 dBi, 0 dBi, and 2 dBi respectively."

What they mean is the system is powered by a +30dBm 1 Watt (!!!) transmitter hooked to a +6dBi antenna. That is a lot of power. Equally we could have a low power radio hooked to a photovoltaic cell or a simple coil that receives power from another powered coil (I built such system experimenting with wireless power transmission when I was 14 - a battery-less wireless audio headset powered by a coil attached to a speaker output of an FM receiver). The bottom line is, it is not a low power system.

Compare that to Bluetooth Smart. A Bluetooth Smart peripheral (e.g. a sensor node) is capable of sending a complete cryptographically secured (encrypted / authenticated) message in 1ms. This is three radio transmissions, one on each advertising channel, each lasting about 350µs. A transmitter consumes 10mA at 1.8V, while active. This totals (rounded up) to 20µJ. Yes 20µJ is the energy required to send a complete digitally signed message (and not just an unsecured beacon).

A coin cell battery is rated at 500mAh / 3.3V. This is 6kJ of energy stored. Enough to transmit 300 million messages. So a Bluetooth Smart transmitter sending a message every 10 seconds will last for 94 years on a coin cell (the calculation does not take into account self-discharging of a battery and some voltage conversion inefficiency, but anyway even half of that number is more than enough).

During that time the 1W "plugged-in device" that powers the passive WiFi system will consume a couple of thousand kWh of energy.

I'm not saying I don't like the idea of passive backscatter wave forming using purely digital approach. It is a great piece of engineering. But those of your who are amazed or scared by the 10000 "breakthrough" number, please do your homework and read the fine print. It may have an influence on your opinion