Breathing/heart rate monitoring with 24GHz radar

[Kasper]

I remember seeing a known router manufacturer talking about this recently and thinking it was pretty far fetched and now seeed says they can do it with a $30 module!? 

Seems like a neat solution for occupancy detection though they say it'll work in hotels and I find it hard to imagine how they'd detect people in one room without false triggers from people in neighboring rooms.

Breathing/heart rate monitoring though, that is hard to believe.  Anyone familiar with this?  Is this feasible?

Just for fun: try to find the poorly soldered cap.

https://www.seeedstudio.com/24GHz-mmWave-Radar-Sensor-Fall-Detection-Module-p-5268.html

[PlainName]

I think it's possible. You basically filter out movement which isn't around 1-2Hz, and if you're good you could derive that 1-2Hz signal even if it's moving around. I would hazard a guess that it's one of those things that looks pretty complicated or tedious to work out but is doable if you had enough paper and time to do it. Which is the kind of things computers are excellent at, so just needs someone to program the algorithm.

[Kleinstein]

The 24 Ghz range is not very good in penerating walls. Of cause it depends - walls in tradional Japanese and tranditional Greenland homes are quite different.
There are relatively cheap radar modules for motion detection / speed measurement - they however give essentially no distance / direction information, just a spectrum of movements from doppler shift.

The are radar instruments to detect heart and breathing movements to detect / search persons, so the principle can work.

[PlainName]

For room occupancy detection, not penetrating walls is probably a good selling point

[Marco]

Seems like a neat solution for occupancy detection though they say it'll work in hotels and I find it hard to imagine how they'd detect people in one room without false triggers from people in neighboring rooms.

It has no penetration.

Breathing/heart rate monitoring though, that is hard to believe.  Anyone familiar with this?  Is this feasible?

I found this paper the last time the subject comes up, it seems plausible it can detect it.

https://www.researchgate.net/profile/Changzhi-Li-3/publication/273396358_A_Hybrid_FMCW-Interferometry_Radar_for_Indoor_Precise_Positioning_and_Versatile_Life_Activity_Monitoring/links/576a534408ae12d1128fb3fe/A-Hybrid-FMCW-Interferometry-Radar-for-Indoor-Precise-Positioning-and-Versatile-Life-Activity-Monitoring.pdf

Though the module in question doesn't seem to have heart rate monitoring, only breathing detection.

[radar_macgyver]

It will work under controlled conditions. One of the issues is the signal characteristics change dramatically with look angle. It can be made to work when looking perpendicular to the chest but doing so for arbitrary angles is a hard task. They do state that breathing detection only works at short ranges. I did this with an older X-band Gunn-diode module without modulation but didn't use FMCW - just plain CW, and looked for the doppler at low frequencies. Same as the experiment posted by @ALW with the NanoVNA, except with the Gunnplexer and a horn. Maybe with FMCW there are other techniques available? With a fine enough range resolution and looking for movement in range rather than doppler?

Does anyone have documentation for the serial data from this sensor? Love that Seeed summarizes version 1.5 of their firmware as "Overall perfection"...

[Kasper]

Seems like a neat solution for occupancy detection though they say it'll work in hotels and I find it hard to imagine how they'd detect people in one room without false triggers from people in neighboring rooms.

It has no penetration.

Breathing/heart rate monitoring though, that is hard to believe.  Anyone familiar with this?  Is this feasible?

I found this paper the last time the subject comes up, it seems plausible it can detect it.

https://www.researchgate.net/profile/Changzhi-Li-3/publication/273396358_A_Hybrid_FMCW-Interferometry_Radar_for_Indoor_Precise_Positioning_and_Versatile_Life_Activity_Monitoring/links/576a534408ae12d1128fb3fe/A-Hybrid-FMCW-Interferometry-Radar-for-Indoor-Precise-Positioning-and-Versatile-Life-Activity-Monitoring.pdf

Though the module in question doesn't seem to have heart rate monitoring, only breathing detection.

They mention breathing/heart rate in the 'scenario' section but leave out heart rate in other sections.

[SiliconWizard]

"Contactless" heart rate monitoring using radars is a real thing. But it only works reliably at relatively short distances. Through walls at a few meters distance? Yeah. Sure.

[Kasper]

"Contactless" heart rate monitoring using radars is a real thing. But it only works reliably at relatively short distances. Through walls at a few meters distance? Yeah. Sure.

That is impressive.  I wonder if this will ever replace the PPGs (LEDs and photodiodes) used in wearables.

[PlainName]

"Contactless" heart rate monitoring using radars is a real thing. But it only works reliably at relatively short distances. Through walls at a few meters distance? Yeah. Sure.

I think you missed the part where they DIDN'T say it works through walls. And, in fact, you wouldn't want it to for room occupancy monitoring. Further, they say:

Measurement distance:

    Motion perception maximum distance: up to 12 meters
    Micromotion perception maximum distance: up to 5 meters
    Body perception maximum distance: up to 3 meters

So the distance stuff isn't specified as what you say it won't do. Did you perhaps just see the thread title and hit reply?

[Marco]

It will work under controlled conditions. One of the issues is the signal characteristics change dramatically with look angle. It can be made to work when looking perpendicular to the chest but doing so for arbitrary angles is a hard task.

It will stop showing up as doppler when the motion is perpendicular, on the other hand it will start showing up as a change in reflection strength for two ranges of distance at the breathing frequency. You should be able to detect both with FMCW and combine the effects in a single measure.