35 ms·

I wonder how cheap and low power we could make tracking beacons if we didn't actually have to send any data and could accept some privacy risk. Imagine like, a

I wonder how cheap and low power we could make tracking beacons if we didn't actually have to send any data and could accept some privacy risk. Imagine like, an airtag, but the ID only needs to be 1 byte long, or even just a simple presence signal. You'd put it on everything you own that's portable and easy to lose. When you want to find it you just search for signals. Maybe it could have an NFC silencing feature. Perhaps it could only activate after 10 minutes of not detecting movement, while also not being near a base station (The base stations would mark a location as the "safe place you won't forget"). So you won't get any interference from anything people are carrying. You won't pick up anything safely stashed where it actually belongs. But you can quickly sweep an area for any lost property (Which could be good or bad, depending on if it's a helpful security guard or someone who believes in finders keepers looking...). You wouldn't ever have to receive any data, except to detect base stations, which would only need maybe a 2ft range. But it wouldn't solve the big issue with losing stuff, detecting when you leave a location without a thing.

lxgr 3y ago
> Imagine like, an airtag, but the ID only needs to be 1 byte long, Well, then you could only ever track 255 different tags :) AirTags are already pretty minimal. Their battery lasts about a year, and I imagine an open source equivalent should be possible to build for under $5. This is purely guesstimated based on the fact that I have a couple of Bluetooth LE thermometer/hygrometers that cost me about $3 a piece that last about a year on the same battery (CR2032) – and these run custom firmware and need to run a display and data logger (to on-chip flash) as well!
- juanuicich 3y ago
  Apparently there are Find My compatible generic tags coming out of China already, and they cost less than $5, shipping included. I say apparently because I’m still waiting for one to arrive and test.
  - lxgr 3y ago
    Interesting! I would have thought that Apple would add some sort of tamper-proof hardware authentication mechanism, given how expensive and privacy-sensitive AirTags are.
    - sroussey 3y ago
      They opened it up to avoid anti-trust issues.
  - r00fus 3y ago
    This sounds awesome - got a link?
    - Guest71022 3y ago
      https://a.aliexpress.com/_mrvl4Xq https://a.aliexpress.com/_mrvl4Xq
      - ornornor 3y ago
        Interesting. Although your link is 11.70$ with 50% off for me, but if you search for “find my Apple” you get many more results with lots of options around 5$
        hrrsn 3y ago
        They seem to come up in the home page deals section quite often for me. I grabbed a couple that were $3.50 a pop, and so far they've behaved exactly like an Airtag (albeit with a worse speaker)
        ornornor 3y ago
        Neat. Are they also water resistant (splashes/rain)?
        petemir 3y ago
- eternityforest 3y ago
  You'd just have to have IDs repeat, and software that could tell the difference based on signal strength. If you have two on the same ID that have different RSSI, show them as two tags. You'd have to do some guesswork though to figure out what's going on but it would be better than nothing, and you might be able to get the size down below what CR2032 allows.
  - lxgr 3y ago
    But why only send a single byte if you can send a full UUID within a rounding error of the same resource usage? If you want it to be compatible with most smartphones, you'll need Bluetooth LE anyway, and that already uses at least 10 bytes on header fields for broadcast packets.
    - eternityforest 3y ago
      I suppose it is effectively a rounding error unless you change the actual protocol. I wouldn't be surprised if there was some trivial way for the manufacturers to add ultra low speed non-bluetooth data to the Bluetooth chips though, so you could have 2400 baud IDs with microwatts of power. But it's probably not worth it since I suspect the real power drain on airtags comes from the high update rates and maybe the 2 way communication.
- imp0cat 3y ago
  A year is a long time, but you can also get a CR2032 to USB converter and use that as a constant power source. Why? If you're using your smart tag to track your car, you can hide it somewhere deep inside and connect it to the vehicle's battery so it always has power. Then you never have to worry about changing batteries.

lxgr 3y ago

> Imagine like, an airtag, but the ID only needs to be 1 byte long, Well, then you could only ever track 255 different tags :) AirTags are already pretty minimal. Their battery lasts about a year, and I imagine an open source equivalent should be possible to build for under $5. This is purely guesstimated based on the fact that I have a couple of Bluetooth LE thermometer/hygrometers that cost me about $3 a piece that last about a year on the same battery (CR2032) – and these run custom firmware and need to run a display and data logger (to on-chip flash) as well!

juanuicich 3y ago
Apparently there are Find My compatible generic tags coming out of China already, and they cost less than $5, shipping included. I say apparently because I’m still waiting for one to arrive and test.
- lxgr 3y ago
  Interesting! I would have thought that Apple would add some sort of tamper-proof hardware authentication mechanism, given how expensive and privacy-sensitive AirTags are.
  - sroussey 3y ago
    They opened it up to avoid anti-trust issues.
- r00fus 3y ago
  This sounds awesome - got a link?
  - Guest71022 3y ago
    https://a.aliexpress.com/_mrvl4Xq https://a.aliexpress.com/_mrvl4Xq
    - ornornor 3y ago
      Interesting. Although your link is 11.70$ with 50% off for me, but if you search for “find my Apple” you get many more results with lots of options around 5$
      - hrrsn 3y ago
        They seem to come up in the home page deals section quite often for me. I grabbed a couple that were $3.50 a pop, and so far they've behaved exactly like an Airtag (albeit with a worse speaker)
        ornornor 3y ago
        Neat. Are they also water resistant (splashes/rain)?
      - petemir 3y ago
eternityforest 3y ago
You'd just have to have IDs repeat, and software that could tell the difference based on signal strength. If you have two on the same ID that have different RSSI, show them as two tags. You'd have to do some guesswork though to figure out what's going on but it would be better than nothing, and you might be able to get the size down below what CR2032 allows.
- lxgr 3y ago
  But why only send a single byte if you can send a full UUID within a rounding error of the same resource usage? If you want it to be compatible with most smartphones, you'll need Bluetooth LE anyway, and that already uses at least 10 bytes on header fields for broadcast packets.
  - eternityforest 3y ago
    I suppose it is effectively a rounding error unless you change the actual protocol. I wouldn't be surprised if there was some trivial way for the manufacturers to add ultra low speed non-bluetooth data to the Bluetooth chips though, so you could have 2400 baud IDs with microwatts of power. But it's probably not worth it since I suspect the real power drain on airtags comes from the high update rates and maybe the 2 way communication.
imp0cat 3y ago
A year is a long time, but you can also get a CR2032 to USB converter and use that as a constant power source. Why? If you're using your smart tag to track your car, you can hide it somewhere deep inside and connect it to the vehicle's battery so it always has power. Then you never have to worry about changing batteries.