12

Bluetooth beacons for localizing lost objects are beginning to spread. You can locate them using an app on your smartphone (yes, and create an account, share on facebook your objects, ...) with a closer/further logic.

I tried one but one day my keys were in fact in my car which was out my house so detecting the beacon from my desk wouldn't work.

I was wondering if Bluetooth is such a good protocol for object location at home, especially for multiple obstacles and outdoor. Other possible networks might be:

So should we stay on BT beacons or can other protocols be more reliable especially for home usage?

  • Are you considering using Bluetooth Low Energy? It might fit your use case better, since the power usage and range characteristics are much better than older versions of Bluetooth. – Aurora0001 Feb 15 '17 at 9:57
  • There is a Bluetooth Low energy tag on my question and I think/suppose all these beacons use LE to avoid pairing and power consumption. My question is: even with LE I can't detect keys 2 meters outside my home, is it the beacon or the protocol which don't work? – Goufalite Feb 15 '17 at 10:02
  • Sorry, didn't see the tag. It seems odd that it failed at such a short range; how much shielding is there between the beacon and the object? A thick wall, perhaps? – Aurora0001 Feb 15 '17 at 12:42
  • 1
    A thick wall and the armature of my car, which might act as a Faraday cage... – Goufalite Feb 15 '17 at 13:03
  • 2
    Most of those beacons are based on your phone always being with you so that it can look back to the last time it had contact with the beacon, and assume that the beacon is still at that last location. Could you specify how large the area you expect to cover is? And to what degree you can set up an infrastructure throughout that area, or do you want a centralized device that can sense anywhere on your (possible large) property? – Wayne Feb 15 '17 at 14:41
8

The problem is essentially one of receivers and of power.

A "find my keys" type of beacon needs to be small enough to attach to your keychain and you probably don't want to have to regularly recharge the beacon (you wouldn't lose your keys if they're always in their charger) so that requires either:

  • a passive technology without a battery.
  • an active technology running of a battery that is extremely power efficient and doesn't require recharging/battery replacement for months or preferably years.

Then you also need a receiver to locate the beacon and in the case of a passive technology you also need a transmitter to power and activate the beacon. The options for receivers/transmitters are either:

  • a dedicated and/or proprietary receiver unit, which allows the manufacturer to select optimised frequencies, protocols and technologies, but which means you need to bring out that receiver when you need to find your keys.
  • use the smartphone which you already carry as the receiver. That makes the solution a lot cheaper, you only have to buy beacons, but also limits the beacon manufacturer to technologies, frequencies and protocols that are commonly implemented on smartphones.

The advantage of passive beacons is that they're likely to be quite reliable and cheap to manufacture, the expense is in a transmitter/receiver combo that works from a reasonable distance. Most likely you'll need special purpose device to take the role of tranmitter/receiver as smartphones currently only support NFC as a passively powered protocol, near-field communication with a range of a couple of cm.
An example that works over significant distances would be the Recco beacons and receivers marketed for finding people buried by avalanches.

For active beacons the main issue is power consumption. As you listed there are a number of different protocols and solutions that are low powered and energy efficient, but since nobody really seems to want to depend on separate receivers but would rather use a smartphone effectively that means either Bluetooth or WiFi.
As currently phones can only be connected to one WiFi network at the time and when operating as a hotspot can't be connected to another WiFI network at all, that is probably not the most desirable protocol.
With Bluetooth 4 and above on the other hand a large number of devices can be connected simultaneously and that standard also comes with the Bluetooth low energy variant. You also get a reasonable distance from Bluetooth.

So should we stay on BT beacons or can other protocols be more reliable especially for home usage?

In summary: unless you are willing to use a specific receiver to find your beacon Bluetooth Low Energy is the best choice.

7

This is an unusual use-case where the data throughput which you require is really low - range and low power are the driving factors. This points in the direction of a low data-rate, narrow band protocol (or maybe one using a spreading code, such as GPS).

Looking for long-range through walls, etc. implies a low RF frequency too, 433 MHz would probably be better than 2.5 GHz.

One drawback with the standard protocols is that they are designed to detect errors in transmission (where they can't be corrected) and discard the data. In fact, you want to use maybe 100 bits of data, and can accept loosing 10 in the noise. I think overall, you have the best chance of making a good design if you set out to develop your own low bitrate, range optimised protocol.

This is probably an impractical solution, but worth being aware of.

  • 1
    With your last sentence, you mean I should implement my own RF protocol? – Goufalite Feb 15 '17 at 9:47
  • 2
    @Goufalite For the best result, yes. This is feasible even using off-the-shelf RF modules, if you consider just the symbol sequences which are transmitted (not the modulation and other transmit parameters). – Sean Houlihane Feb 15 '17 at 9:51
  • No offence to the OP, but do you really believe that he is more capable than the professionals at big multinationals, or has anything like their resources, both in terms of finance and of equipment, access to other professionals, etc., etc, etc? I guess you are talking of Layer one of the OSI model here? He will, of course, have to manufacture both ends of his system, and spend a great deal of time debugging. In short, OP, I would not advise this course. Unlesshe is looking for a long term hobby, which might consume a lot of time & monye & will be more expesive than a COTS solution. – Mawg Feb 16 '17 at 8:22
  • I never really understood the model, but more like layer 2 and 3. Layer1 is only really hard because of regulations. All I'm saying is COTS is unlikely to address this niche with a full stack. I do suggest that the OP may not have the relevant experience, but it can be learnt. Also, the OP needs to know if an expert is able to contribute to the field and I'm saying yes, probably not many people have taken a recent commercial interest in this. Was about 20 years ago when I was asked to investigate it... – Sean Houlihane Feb 16 '17 at 8:34
6

I would like to say that Yes, Bluetooth is a good candidate for indoor positioning.

Depending on how good positioning you want, of course.

If you have Bluetooth beacons scattered over your house it will probably be as simple as cell positioning - the beacon that last saw your key chain will be in the same room as the key chain. With one beacon per room, it will be simple to find out where it is.

There are actually quite a few indoor positioning solutions out there that relies on bluetooth.

  • For a "Bluetooth beacon" to "see" your key chain, one of them would have to act as a central, not a peripheral. Existing beacons tend to act as peripherals so that your phone can see them. – Dan Hulme Feb 27 '17 at 10:23
  • I upvoted your answer, but would love to see you expand on "There are actually quite a few indoor positioning solutions out there that relies on bluetooth" – Mawg Apr 29 at 9:49
3

If you are an experienced tinkerer, you might consider Goufalite’s solution.

Personally, I think that market forces drive the commercially available solutions to be the best that they can currently be, but feel free to continue Googling for an answer to your question.

With regard to some of the technologies that you mention:

  • Bluetooth - is by far the most common application in this field. Occam’s razor would lead us to believe that there are sound technical & commercial reasons for this.
  • Wi-Fi – you say is energy hungry. BUT, if you only use that energy when looking for a lost item, is that really a major consideration
  • RFID readers and passive pages are, as you say, cheap – up to about two inches, then things start to get expensive.
  • I know nothing of Z-wave or LoRa, but you don’t seem keen on them, so, but a process of elimination, it would seem to be either rBT or Wi-Fi.

One thing that you do not make clear, is whether you are considering building a solution, or only wish to purchase.

My answer is not to your question, but to the problem which eventuated it. Reverse your way of looking at it. Instead of “I’ve lost my keys and need to find them”, look for a solution which informs you when your keys are out of range. These are ubiquitous and cheap and the Googling thereof is left as an exercise for the reader.

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Not the answer you're looking for? Browse other questions tagged or ask your own question.