These sensors are becoming increasingly popular in the home automation space, replacing traditional PIR sensors because, unlike PIR sensors they can still detect someone even if they're sitting relatively still (just breathing can be enough to trigger one!). I just wish my office would add them, the lights are always turning off on me.
FMCW radar is great for precise detection, and TFA does a good job of describing it.
It would do an even better job if it also mentioned that an FMCW based human presence detector consumes several orders of magnitude more power than a PIR sensor, which means battery operation will either make the sensor very large or require frequent battery changes.
Precise? May depend on usage. I worked at a company that used them for measuring traffic, but the speed detection was not good. Vehicles have multiple reflecting surfaces, and it's hard to interpret multiple reflections.
The fact that apparently, this tiny little sensor board can reliably measure your heart rate from a distance by using radar is kind of blowing my mind.
Sensing features are scheduled for 2024 consumer Wi-Fi 7 routers, with through-wall coverage (2.4Ghz) or high-resolution within a room (802.11ay 60Ghz mmWave), https://news.ycombinator.com/item?id=34480760
Heart rate? Go a step further and consider you can track discrete people with these things. Like walk in front of one and it can build a unique tracking profile of you and track you as an individual. The profiles (I've been told) can be as unique as a fingerprint.
That's not completely false though. These radars have been used lately in Robotics as ways to provide speed measures in localization and Mapping algorithms (SLAM, or lower-level algorithms like VIO + radar). Providing a speed measure on top of images and IMU often greatly increases the precision of the result. Some of these algorithms are actually Deep Neural Network, so you could argue that these have been 'adopted' by AI.
I'm confused how the Doppler shift is measured in the case of the continuously modulated frequency? It seems both range and Doppler shift are encoded in the reflected signal - how are they separated?
Edit: or is it the case that a full period of the reflected signal must be detected first before Doppler can be calculated?
"Modulation can be turned off on alternate scans to identify velocity using unmodulated carrier frequency shift. This allows range and velocity to be found with just one radar set."
Doppler is measured across multiple chirps or pulses. So doppler is not really encoded in the same signal used for range. Relative motion induces a phase rotation across multiple chirps, which can be coherently combined with a Fourier transform.
You're thinking sawtooth VCO frequency variation because its the simplest way to teach how distance ranging works with FMCW
The "real world" radars usually use a triangle wave. If no movement, the delta on rising and falling parts of the triangle are the same. A doppler will shift the height so to speak so the delta output instead of being constant (distance) will have a square wave look to it and the height of that sq wave will be the doppler component.
So pseudocode looks something like figure out the average delta when the triangle modulation is going up, then average when its going down, and the average of the averages is the distance and the difference of the averages is the doppler.
Note that "real radars" and "military radars" use wildly more interesting modulation schemes. Sawtooths and triangles were cutting edge in 1950 not so much anymore. Imagine if you generated a known pseudo-random or crypto-generated digital signal and transmitted that and the remembered what you transmitted for awhile and correlated your known previous values with what you rx a little later; now that was cutting edge in the 60s/70s and thats quite hard to jam compared to legacy methods.
Also I "know" the NWS WSR-88 weather radars transmit some really cool and interesting waveforms for weather experiments although I don't have the details.
Edited: Its too early in the morning for me to think about this, I may have the above backwards. The general concept, anyway, is the rate of change of your triangle wave is way faster than the change in movement, and over the very short term of a couple cycles of the triangle modulation, the returned freq from doppler is a "constant" offset high or low added to the shift from distance. Now during the upslope of a triangle the returned signal will lag and return a lower/negative voltage and during the downslope of a triangle the returned signal will lag and return a higher/positive voltage from the past. So yeah I'm sleepy and I got it backwards above; leaving for the LOLs.
Edited x2: Ugh I'm sleepy. This is like parody leetcode for EEs. Just do a numerical simulation using small simple numbers of modulating a FMCW with a triangle wave (not sawtooth...) and it'll all eventually make sense.
They seem to essentially be "water" detectors. With human bodies being mostly water, they serve as presence detectors almost as a happy side effect. With some of the cheaper/less tuned sensors, they will detect a 5 gallon bucket of water, or a burbling water cooler as a person.
You can get range estimates from a single unit, but direction is poor to non-existent.
I don't see these being good for what you are describing.
Can one of these be used to measure the speed of a golf ball? What is the output signal of a typical module? Do we get just a single float representing speed?
It comes down to power supply and distance, you could, but you would probably need numerous sensors and would probably be better off with a longer wave and bigger power source
PIR sensors are faster than these, though. It makes a fair bit of difference in how long the lights take to turn on when you walk in a room, so some systems use both PIR (for when someone walks in), and mmWave (for when someone is in/leaves).
These radar sensors work because the signal is reflected from the surface of our bodies and doesn't penetrate. The most harmful things penetrate our body, even lead pipes wont hurt you until you drink the contaminated water.
What you're missing is range scales on square of power (or is it twice that for radar? I would think so) and all known forms of radiation have bio effects that scale by linear dose.
So the 60 GHz presence detector transmits at +6 dBmW aka 4 milliwatts or 4 thousandths of a watt. As a side note microwave RF has never been known for power efficiency and I've built many a MMIC amplifier running under 1 percent efficient LOL. Something like a patriot missile AN/MPQ-53 radar transmits at 10 kilowatts, which is 2.5 million times the power of these detectors. If you want to burn thru radar jamming at 350 miles or whatever the unclassified spec is for the -53 radar, its going to take some serious wattage. Of course that's nothing compared to an old fashioned UHF broadcast transmitter or even a modern FM broadcast radio transmitter.
Anyway the point of the above is in theory if RF caused cancer or death or covid or whatever its being blamed for now, the "medical signal" would be 2.5 million times stronger in military anti-aircraft units than in rando presence detector general public, to a first estimate. Seeing as there's no difference in death rates for ADA soldiers vs artillery or any other army branch that I'm aware of, and the rate would be 2.5 million times lower than... nothing... for general public using presence detectors, it seems unlikely that even massive scaling up to a billion users would have a measurable effect.
Similar comedic examples can be generated for broadcast engineers. I'm friends with a couple broadcast EEs and old broadcast engineers just don't die (LOL). If a cellphone (100 times the power of a presence detector) gave even one human being per year early cancer, your average broadcast engineer would be dead in their first five minutes of work or something like that because they're always screwing around with enormous power levels, no matter if its permitted or not technically.
The thing about electronics, is much like radiation, we can generate and measure sources over 9, maybe 12 orders of magnitude every day all over the planet.
These sensors are becoming increasingly popular in the home automation space, replacing traditional PIR sensors because, unlike PIR sensors they can still detect someone even if they're sitting relatively still (just breathing can be enough to trigger one!). I just wish my office would add them, the lights are always turning off on me.
At my office I have a "Drinking Bird" toy I set up near the motion sensor when I need the lights to stay on.
If I forget to take it down is self-limiting because eventually the dish of water evaporates.
FMCW radar is great for precise detection, and TFA does a good job of describing it.
It would do an even better job if it also mentioned that an FMCW based human presence detector consumes several orders of magnitude more power than a PIR sensor, which means battery operation will either make the sensor very large or require frequent battery changes.
Precise? May depend on usage. I worked at a company that used them for measuring traffic, but the speed detection was not good. Vehicles have multiple reflecting surfaces, and it's hard to interpret multiple reflections.
The fact that apparently, this tiny little sensor board can reliably measure your heart rate from a distance by using radar is kind of blowing my mind.
Sensing features are scheduled for 2024 consumer Wi-Fi 7 routers, with through-wall coverage (2.4Ghz) or high-resolution within a room (802.11ay 60Ghz mmWave), https://news.ycombinator.com/item?id=34480760
Heart rate? Go a step further and consider you can track discrete people with these things. Like walk in front of one and it can build a unique tracking profile of you and track you as an individual. The profiles (I've been told) can be as unique as a fingerprint.
Any camera can do that.
Fingerprints tend not to change. Bad shoes, ankle issues, back aches, carrying things, age have all effected my gait.
Title needs changing.....
......I can't be the only one wondering what mega metre radar was.
That's a cute little set of low-cost boards.
The copy, though... "Radar has been adopted as a simple and practical solution in Artificial Intelligence."?
That's not completely false though. These radars have been used lately in Robotics as ways to provide speed measures in localization and Mapping algorithms (SLAM, or lower-level algorithms like VIO + radar). Providing a speed measure on top of images and IMU often greatly increases the precision of the result. Some of these algorithms are actually Deep Neural Network, so you could argue that these have been 'adopted' by AI.
Sounds like more of a robotics solution than an AI solution.
Not those units, though. They're non-directional.
Papers on the use of machine learning in Device-Free Wireless Sensing (DFWS) with Wi-Fi, https://scholar.google.com/scholar?hl=en&as_sdt=5%2C44&sciod...
I'm confused how the Doppler shift is measured in the case of the continuously modulated frequency? It seems both range and Doppler shift are encoded in the reflected signal - how are they separated?
Edit: or is it the case that a full period of the reflected signal must be detected first before Doppler can be calculated?
"Modulation can be turned off on alternate scans to identify velocity using unmodulated carrier frequency shift. This allows range and velocity to be found with just one radar set."
quote from https://navtechradar.com/explore/fmcw-radar/#:~:text=Frequen....
I believe this is where the "Ambiguity Function" comes in. For a given waveform, it tells you the ambiguity between Doppler shift and time delay.
Doppler is measured across multiple chirps or pulses. So doppler is not really encoded in the same signal used for range. Relative motion induces a phase rotation across multiple chirps, which can be coherently combined with a Fourier transform.
You're thinking sawtooth VCO frequency variation because its the simplest way to teach how distance ranging works with FMCW
The "real world" radars usually use a triangle wave. If no movement, the delta on rising and falling parts of the triangle are the same. A doppler will shift the height so to speak so the delta output instead of being constant (distance) will have a square wave look to it and the height of that sq wave will be the doppler component.
So pseudocode looks something like figure out the average delta when the triangle modulation is going up, then average when its going down, and the average of the averages is the distance and the difference of the averages is the doppler.
Note that "real radars" and "military radars" use wildly more interesting modulation schemes. Sawtooths and triangles were cutting edge in 1950 not so much anymore. Imagine if you generated a known pseudo-random or crypto-generated digital signal and transmitted that and the remembered what you transmitted for awhile and correlated your known previous values with what you rx a little later; now that was cutting edge in the 60s/70s and thats quite hard to jam compared to legacy methods.
Also I "know" the NWS WSR-88 weather radars transmit some really cool and interesting waveforms for weather experiments although I don't have the details.
Edited: Its too early in the morning for me to think about this, I may have the above backwards. The general concept, anyway, is the rate of change of your triangle wave is way faster than the change in movement, and over the very short term of a couple cycles of the triangle modulation, the returned freq from doppler is a "constant" offset high or low added to the shift from distance. Now during the upslope of a triangle the returned signal will lag and return a lower/negative voltage and during the downslope of a triangle the returned signal will lag and return a higher/positive voltage from the past. So yeah I'm sleepy and I got it backwards above; leaving for the LOLs.
Edited x2: Ugh I'm sleepy. This is like parody leetcode for EEs. Just do a numerical simulation using small simple numbers of modulating a FMCW with a triangle wave (not sawtooth...) and it'll all eventually make sense.
Anyone here know if it can be used for generating 3D surface profiles? Or material thickness measurements
Based on my experiences with them, no.
They seem to essentially be "water" detectors. With human bodies being mostly water, they serve as presence detectors almost as a happy side effect. With some of the cheaper/less tuned sensors, they will detect a 5 gallon bucket of water, or a burbling water cooler as a person.
You can get range estimates from a single unit, but direction is poor to non-existent.
I don't see these being good for what you are describing.
Thank you.
Wondering if anyone here would know this..
Can one of these be used to measure the speed of a golf ball? What is the output signal of a typical module? Do we get just a single float representing speed?
Ainstein demonstrated this with tiger woods. The raw output is very dense and contains range and Doppler information.
The first order effect (distance) is the difference frequency. The second order effect (velocity) is the change in frequency.
Noise (SNR, distance to the ball, aperture) will determine how well this works, since too much integration could muddy the signals.
It’s worth a try! Good luck.
It comes down to power supply and distance, you could, but you would probably need numerous sensors and would probably be better off with a longer wave and bigger power source
I note the comparison to the common PIR motion sensors which are affordable but not that good.
PIR sensors are faster than these, though. It makes a fair bit of difference in how long the lights take to turn on when you walk in a room, so some systems use both PIR (for when someone walks in), and mmWave (for when someone is in/leaves).
I think this much radio waves around us constantly will end up being the lead pipes of our generation
ya think the new version of the song will be:
"Radio killed the Video Star" ?
cf https://www.youtube.com/watch?v=W8r-tXRLazs
These radar sensors work because the signal is reflected from the surface of our bodies and doesn't penetrate. The most harmful things penetrate our body, even lead pipes wont hurt you until you drink the contaminated water.
I’m pretty sure a lead pipe will hurt you if it penetrates your body. (Sorry couldn’t resist)
A copper pipe would too :P
Your description fits perfectly for a perfect world. But there is no 100% reflection in a real world.
So it depends on the wavelength, but no, perfect reflection does not happen.
Further, as an IOT engineer, I can tell you we do not currently have enough data to be sure
I’m getting downvoted by people who probably aren’t even in the signals field lol…
Remember just 70 years ago when “lobotomies were sound science and totally safe”
What you're missing is range scales on square of power (or is it twice that for radar? I would think so) and all known forms of radiation have bio effects that scale by linear dose.
So the 60 GHz presence detector transmits at +6 dBmW aka 4 milliwatts or 4 thousandths of a watt. As a side note microwave RF has never been known for power efficiency and I've built many a MMIC amplifier running under 1 percent efficient LOL. Something like a patriot missile AN/MPQ-53 radar transmits at 10 kilowatts, which is 2.5 million times the power of these detectors. If you want to burn thru radar jamming at 350 miles or whatever the unclassified spec is for the -53 radar, its going to take some serious wattage. Of course that's nothing compared to an old fashioned UHF broadcast transmitter or even a modern FM broadcast radio transmitter.
Anyway the point of the above is in theory if RF caused cancer or death or covid or whatever its being blamed for now, the "medical signal" would be 2.5 million times stronger in military anti-aircraft units than in rando presence detector general public, to a first estimate. Seeing as there's no difference in death rates for ADA soldiers vs artillery or any other army branch that I'm aware of, and the rate would be 2.5 million times lower than... nothing... for general public using presence detectors, it seems unlikely that even massive scaling up to a billion users would have a measurable effect.
Similar comedic examples can be generated for broadcast engineers. I'm friends with a couple broadcast EEs and old broadcast engineers just don't die (LOL). If a cellphone (100 times the power of a presence detector) gave even one human being per year early cancer, your average broadcast engineer would be dead in their first five minutes of work or something like that because they're always screwing around with enormous power levels, no matter if its permitted or not technically.
The thing about electronics, is much like radiation, we can generate and measure sources over 9, maybe 12 orders of magnitude every day all over the planet.