Maker Pro

Wi-Fi IoT using RF backscatter

Is backscatter Wi-Fi a practical solution for IoT products?


  • Total voters
    4
  • Poll closed .

hevans1944

Hop - AC8NS
Jun 21, 2012
4,878
Joined
Jun 21, 2012
Messages
4,878
Recently someone posted a video showing an IoT (Internet of Things) device that did not require an RF transceiver to link into a local Wi-Fi network. Instead, it modulated the antenna impedance digitally with a MOSFET to utilize RF backscatter from a local, powered, wireless Wi-Fi transceiver to piggyback low-bandwidth data packets to a LAN Wi-Fi router.

You still need power for the digital electronics in the IoT device that is modulating the backscattered RF, but there is no power required to operate an RF transceiver in the IoT device. All this was invented about two years ago by students at the University of Washington, and they now have a commercial entity to exploit the technology.

I would like to open a thread and try to determine if this technology can be used by highly qualified hobbyists for home use. There may be practical limitations on range and data bandwidth, but it sounds exciting... at least to me.

Here is a link to get the ball rolling. There are papers published on the subject, but Google is your friend here in finding more of them.
 

AnalogKid

Jun 10, 2015
2,884
Joined
Jun 10, 2015
Messages
2,884
It's real enough. Glad to see it finally coming to life after only 50 *years* of development. If you think about it, it's not that different in concept from in-ground traffic light sensors, except that the cars are going by at 54 million per second and its detecting the difference between a Beetle and an SUV.

ak
 

hevans1944

Hop - AC8NS
Jun 21, 2012
4,878
Joined
Jun 21, 2012
Messages
4,878
I think perhaps you miss the point. This "new" technology provides a way to implement low-power IoT devices passively, without the huge power overhead of a Wi-Fi transceiver. One possible goal is to harvest wireless power to provide the minuscule power requirements of a passive IoT device, but that is not the primary goal. The primary goal is to reduce power requirements of IoT devices from hundreds of milliwatts to a few dozen microwatts by completely eliminating the need for an RF transceiver. I think this is considerably different than how an in-ground traffic light sensor operates. However, if I am wrong about that, please enlighten me.

A local, line-powered, "master" Wi-Fi transceiver emits a single-frequency tone on a frequency adjacent to a single wi-fi channel. A local IoT device is a passive receiver of this tone but can modify the frequency and add information to radiation backscattered from its antenna by modulating the antenna impedance with a MOSFET switch. All of this is low-power CMOS logic implemented with a FPGA device drawing microwatts of power from the IoT's power source. It uses this capability to create modulated backscattered radiation on a nearby adjacent Wi-Fi channel, establishing a Wi-Fi link to the line-powered Wi-Fi. From there it gets a lot more complicated, but the IoT device eventually becomes a part of the local area network serviced by a nearby wireless router that provides Internet connectivity. Ranges up to a couple dozen feet or so, through intervening drywall-covered walls, at IEEE-802.11b rates have been demonstrated between a passive IoT device and an active device such as a cell phone. See this recent paper.

The implication is you can use your cell phone and a local area network to communicate with IoT devices installed at various places in your home. The IoT device would typically be battery operated and does not need to harvest wireless power, although that could be an option for very low power-consuming devices such as a programmable thermostat for example. A security camera would probably require more power, and might even need line-power to operate pan/tilt and variable focus mechanisms, but it would not need the power required by an active Wi-Fi transceiver. Similarly, a refrigerator or freezer high-temperature or door-open alarm would certainly have power available along with a battery back-up against a circuit-breaker trip, but it would not need the overhead baggage of a full-blown Wi-Fi transceiver.

There is a private firm licensed to exploit this technology, but so far there is very little information at their website other than links to pages reporting on the technology can be found there. My question is: can the hobbyist successfully implement a passive IoT device using off-the-shelf technology? The FPGA programming might be a bit daunting. I've certainly never ventured into that realm, but perhaps there are alternatives. There may also be patents to be considered for commercial development, but it is my understanding that any U.S. citizen can copy a patented invention for their personal use without paying royalties or otherwise infringing on the rights of the patent holder. Someone please tell me if I am wrong about that too.

Hop
 

Gryd3

Jun 25, 2014
4,098
Joined
Jun 25, 2014
Messages
4,098
I think perhaps you miss the point. This "new" technology provides a way to implement low-power IoT devices passively, without the huge power overhead of a Wi-Fi transceiver. One possible goal is to harvest wireless power to provide the minuscule power requirements of a passive IoT device, but that is not the primary goal. The primary goal is to reduce power requirements of IoT devices from hundreds of milliwatts to a few dozen microwatts by completely eliminating the need for an RF transceiver. I think this is considerably different than how an in-ground traffic light sensor operates. However, if I am wrong about that, please enlighten me.

A local, line-powered, "master" Wi-Fi transceiver emits a single-frequency tone on a frequency adjacent to a single wi-fi channel. A local IoT device is a passive receiver of this tone but can modify the frequency and add information to radiation backscattered from its antenna by modulating the antenna impedance with a MOSFET switch. All of this is low-power CMOS logic implemented with a FPGA device drawing microwatts of power from the IoT's power source. It uses this capability to create modulated backscattered radiation on a nearby adjacent Wi-Fi channel, establishing a Wi-Fi link to the line-powered Wi-Fi. From there it gets a lot more complicated, but the IoT device eventually becomes a part of the local area network serviced by a nearby wireless router that provides Internet connectivity. Ranges up to a couple dozen feet or so, through intervening drywall-covered walls, at IEEE-802.11b rates have been demonstrated between a passive IoT device and an active device such as a cell phone. See this recent paper.

The implication is you can use your cell phone and a local area network to communicate with IoT devices installed at various places in your home. The IoT device would typically be battery operated and does not need to harvest wireless power, although that could be an option for very low power-consuming devices such as a programmable thermostat for example. A security camera would probably require more power, and might even need line-power to operate pan/tilt and variable focus mechanisms, but it would not need the power required by an active Wi-Fi transceiver. Similarly, a refrigerator or freezer high-temperature or door-open alarm would certainly have power available along with a battery back-up against a circuit-breaker trip, but it would not need the overhead baggage of a full-blown Wi-Fi transceiver.

There is a private firm licensed to exploit this technology, but so far there is very little information at their website other than links to pages reporting on the technology can be found there. My question is: can the hobbyist successfully implement a passive IoT device using off-the-shelf technology? The FPGA programming might be a bit daunting. I've certainly never ventured into that realm, but perhaps there are alternatives. There may also be patents to be considered for commercial development, but it is my understanding that any U.S. citizen can copy a patented invention for their personal use without paying royalties or otherwise infringing on the rights of the patent holder. Someone please tell me if I am wrong about that too.

Hop
If this tech can be applied to WiFi, I'm sure it can be applied to almost any other RF based technology...
I wonder what kind of battery life we can get out of our phones if they started routing voice through Passive WiFi when you are in a home or business with the appropriate hardware.
I can imaging that the power required to do this with cellular radio would be incredibly high though...
 

AnalogKid

Jun 10, 2015
2,884
Joined
Jun 10, 2015
Messages
2,884
I think perhaps you miss the point. This "new" technology provides a way to implement low-power IoT devices passively, without the huge power overhead of a Wi-Fi transceiver. One possible goal is to harvest wireless power to provide the minuscule power requirements of a passive IoT device, but that is not the primary goal. The primary goal is to reduce power requirements of IoT devices from hundreds of milliwatts to a few dozen microwatts by completely eliminating the need for an RF transceiver. I think this is considerably different than how an in-ground traffic light sensor operates. However, if I am wrong about that, please enlighten me.

I get the point. In fact, on another forum I've struggled to explain it on an someone who keeps talking about how a 100 uW receiver cannot power a 100 mW transmitter.

The traffic light sensor isn't a perfect analogy, but it covers the basics. The coil in the ground is the L of an L-C tank oscillator. When something metallic and sufficiently large approaches the soil, the inductance changes and so the frequency changes,and the frequency shift is detected and acted upon. Not quite the same as backscatter, but close enough - create a "disturbance in the force" by disturbing an existing magnetic field, and the disturbance can be detected at some distance. Another imperfect analogy is the electric guitar, although for that it is a DC magnetic field being disturbed.

As for deriving operating power from the field being modulated, many RFID tags do that, as does my ID card. But the range is 1 inch for the ID card, and not much more for most tags. I don't think true self-powered systems ever will reach feet of range because there just isn't enough energy to harvest with any reasonably-sized antenna.

ak
 

hevans1944

Hop - AC8NS
Jun 21, 2012
4,878
Joined
Jun 21, 2012
Messages
4,878
... I don't think true self-powered systems ever will reach feet of range because there just isn't enough energy to harvest with any reasonably-sized antenna.

ak
I agree with that given the current technology. The power consumption would have to reduced to femto-watt levels, maybe lower if antenna size shrinks in proportion to circuit size. The only application I see on the near horizon is microscopic "bugs" that "phone home" when properly triggered by a coded RF signal. I think these already exist, but not for consumer use.
 

(*steve*)

¡sǝpodᴉʇuɐ ǝɥʇ ɹɐǝɥd
Moderator
Jan 21, 2010
25,510
Joined
Jan 21, 2010
Messages
25,510
That is just so damn cool.
 

AnalogKid

Jun 10, 2015
2,884
Joined
Jun 10, 2015
Messages
2,884
Completely agree. And the tech has been laying around for decades, just waiting for the components to catch up.

Many moons ago I designed a galvanically isolated switch interface for a harsh environment. Squarewave oscillator drives a small modem transformer through a 1K resistor. The secondary of the transformer goes off to the switch; just a switch, no current-limiting resistor, no nothing. When you close the switch, the load on the secondary reflects back to the primary, the amplitude of the squarewave at the resistor/primary junction changes, and this is detected with a comparator. Kilovolt isolated dry circuit interface for pennies, using the same "disturbance in the force" idea.

ak
 

hevans1944

Hop - AC8NS
Jun 21, 2012
4,878
Joined
Jun 21, 2012
Messages
4,878
I haven't done a patent search to see what the authors think is patentable by exploiting basic physics. Probably the software they ginned up is patentable, but who knows what you can slip past a patent examiner these days? Wish I had the resources to pursue this. Oh, well, near-field communication between two adjacent cell phones is enough majic for me, plus it's off-the-shelf right now. As the @AnalogKid said, this ain't exactly new technology, just a clever exploitation of existing technology. Didn't the U.S. once have a gigantic over-the-horizon backscatter RADAR installation for detecting pesky Rusky missiles comin' at us? If they fire that up again, maybe folks in Canada can get free Wi-Fi service.:p
 
Top