Have a look at Hi-Link from China (on Aliexpress).
They offer 50GHz, 10GHz, 24GHz and 60GHz FMCW radar module boards for 10-30€ which are easy to reverse engineer on firmware and PCB level.
A lot of them are CW radars, a few FMCW and they also use the Infineon 60GHz radar chips. Very unusual for Infineon: with all tools and datasheets available without signing a NDA. Down to the register level.
There's an interesting Chinese company around (ICLink with their ICL1122/ICL1112) which offer highly integrated radar ICs. They can spit out raw ADC measurements of their downmixed baseband using Quad SPI at 20 or 40MHz when switching on their debug mode. Price range: ~10€. But datasheets are difficult to find. Example board: LD2410S.
The analog ones are easy to play with. You just need a DAC to drive their VCO and then can sample the I/Q pins with an ADC. That how a lot of the 24GHz modules (like the LD1125H) work.
> The analog ones are easy to play with. You just need a DAC to drive their VCO and then can sample the I/Q pins with an ADC
Do you have any reference or notes on how to access the IQ pins on one of these devices (ideally one of the FMCW ones)? I've been wanting to play around with one of these 24 GHz or 60 GHz units for coherent radar but it seems like most of the boards only report on distances over serial links. If there's an easy way to tap into the analog IF signal after down conversion I'd love to see how to do that!
What's cute about this is how far they went to make hobbyists happy. There's a way to connect it to Bluetooth and WiFi, and an Android app. There's compatibility with some Adafruit products. That's unusual for IC data sheets.
This is a phased array device. Angular resolution 20 degrees, range resolution 1 meter.
It's not a Doppler radar, so it can detect fixed objects. So if you're using it for people detection, you have to tell it where the fixed objects are. A ceiling mounted unit will see the floor. OK for people counting and such. Range is only 10 meters.
If you just want a motion detector to turn on a light, and IR isn't working for you, there are cheaper microwave detectors.
> What's cute about this is how far they went to make hobbyists happy.
Professionals are using hobbyist tools more and more, since these tools are just so much easier to use and don't require wading through datasheets which is like filling out tax forms.
Also IoT usecases require flexibility in connectivity.
Plus, rapid prototyping makes it easier / more likely to bring a volume product to market, and selling the same parts to hobbyists and manufacturers amortizes R&D costs more widely. It’s just makes a ton of sense all the way around.
As far as seeing and ignoring fixed objects, you can also remove any returns that have a near-zero velocity in radar and focus only on those objects that are moving.
Of course, indoor settings have a lot of non-stationary objects as well that might not be targets of interest to you, like fans, curtains blowing in the breeze, etc. So you can also develop algorithms to remove those signatures too.
Seeing fixed objects can be beneficial as well, for example, if you have a sensor deployed in a room but you don't know a priori what the room looks like. Longitudinal results and long range statistics can take you pretty far in seeing the room extents and layout and furniture, etc. Though a lidar sweep is better if you can get it
I’ve played around with this (the KIT-CSK-BGT60TR13C). It is a very interesting gadget and the kit doesn’t need much configuration. In an afternoon I made a car detector for my bike that detects the presence of an approaching car from behind. I thought I had a winning startup idea then found out this idea has existed for years. :(
They don’t suck. The Garmin one is also a rear light. It integrates with the bike computer to show detections.
When a car approaches from behind you get a visualisation on the bike computer showing you how far away the car is and an audio alert. If the car is approaching fast you get a different audio alert and the light flashes differently to warn the approaching car. It also detects up to three cars.
There might be improvements to be had but the implementation is pretty solid.
Pleasantly surprised to see electronics parts manufacturers on front page HN. The advancement of sensors nowadays is staggering - I'm reminded of the recent airborne dust sensor that's literally orders of magnitude smaller because it uses a new backscatter detection mechanism. All this technology available at very reasonable prices too.
It is fascinating to me how 'hackers' went from being people who operated at the margins of established technology (2600, Capn Crunch anyone?), to being the literal establishment, in the sense that probably a good chunk of the 'hackers' here work at one of the big 5 or plausibly could.
> Regulatory issues were likely a major factor that led to the demise of Soli and Motion Sense on future Pixel models. Soli operates in the 60GHz frequency, which is reserved for military and government use in India.. Many of the Google Pixel 4's Motion Sense gestures are available.. Nest Hub's Soli radar extends far enough to detect when you're sleeping, and to track your breathing.
I got one of those cheap waveshare 60GHz modules with heart rate detection a few years ago. My goal was to build a sleep tracker - stick it under the bed pointing up, detect getting in and out of bed and falling asleep.
Unfortunately even just pointed straight at me with no obstructions and nobody else in the room the data was more or less random noise...
I wonder if now (after few more years of development, more reputable manufacturer and more money) it is worth trying again?
Thinking of starting a project to detect rainfall/snow/mist - has anybody used those mmWave sensors in a similar application? I've researched some of those and seems many are build for specific application, like a car rain sensor, which hardly provides broader output to apply further modeling.
Readit News
They offer 50GHz, 10GHz, 24GHz and 60GHz FMCW radar module boards for 10-30€ which are easy to reverse engineer on firmware and PCB level.
A lot of them are CW radars, a few FMCW and they also use the Infineon 60GHz radar chips. Very unusual for Infineon: with all tools and datasheets available without signing a NDA. Down to the register level.
There's an interesting Chinese company around (ICLink with their ICL1122/ICL1112) which offer highly integrated radar ICs. They can spit out raw ADC measurements of their downmixed baseband using Quad SPI at 20 or 40MHz when switching on their debug mode. Price range: ~10€. But datasheets are difficult to find. Example board: LD2410S.
The analog ones are easy to play with. You just need a DAC to drive their VCO and then can sample the I/Q pins with an ADC. That how a lot of the 24GHz modules (like the LD1125H) work.
Do you have any reference or notes on how to access the IQ pins on one of these devices (ideally one of the FMCW ones)? I've been wanting to play around with one of these 24 GHz or 60 GHz units for coherent radar but it seems like most of the boards only report on distances over serial links. If there's an easy way to tap into the analog IF signal after down conversion I'd love to see how to do that!
Deleted Comment
This is a phased array device. Angular resolution 20 degrees, range resolution 1 meter. It's not a Doppler radar, so it can detect fixed objects. So if you're using it for people detection, you have to tell it where the fixed objects are. A ceiling mounted unit will see the floor. OK for people counting and such. Range is only 10 meters.
If you just want a motion detector to turn on a light, and IR isn't working for you, there are cheaper microwave detectors.
Professionals are using hobbyist tools more and more, since these tools are just so much easier to use and don't require wading through datasheets which is like filling out tax forms.
Also IoT usecases require flexibility in connectivity.
Of course, indoor settings have a lot of non-stationary objects as well that might not be targets of interest to you, like fans, curtains blowing in the breeze, etc. So you can also develop algorithms to remove those signatures too.
Seeing fixed objects can be beneficial as well, for example, if you have a sensor deployed in a room but you don't know a priori what the room looks like. Longitudinal results and long range statistics can take you pretty far in seeing the room extents and layout and furniture, etc. Though a lidar sweep is better if you can get it
When a car approaches from behind you get a visualisation on the bike computer showing you how far away the car is and an audio alert. If the car is approaching fast you get a different audio alert and the light flashes differently to warn the approaching car. It also detects up to three cars.
There might be improvements to be had but the implementation is pretty solid.
Dead Comment
"Inside a $1 radar motion sensor" (2024), 100 comments, https://news.ycombinator.com/item?id=40834349
"WhoFi: Deep Person Re-Identification via Wi-Fi Channel Signal Encoding" (2025), https://news.ycombinator.com/item?id=44685869
https://en.wikipedia.org/wiki/Pixel_4#Motion_Sense
> Regulatory issues were likely a major factor that led to the demise of Soli and Motion Sense on future Pixel models. Soli operates in the 60GHz frequency, which is reserved for military and government use in India.. Many of the Google Pixel 4's Motion Sense gestures are available.. Nest Hub's Soli radar extends far enough to detect when you're sleeping, and to track your breathing.
Unfortunately even just pointed straight at me with no obstructions and nobody else in the room the data was more or less random noise...
I wonder if now (after few more years of development, more reputable manufacturer and more money) it is worth trying again?
What steps did you take to limit the detection bandwidth? SNR is often a bandwidth problem, not a sensitivity problem.