Mobile Presence and iBeacon (discussion beginning 2015)

Really nice stuff! Using the app to POST to SmartThings is definitely the best method. I have a strong feeling that ST will have some beacon integration inside the app in the near future as ST already has the Bluetooth permission, at least for Android.

Some brain dump comments below:


  1. iBeacons should be both fast and reliable for geofence detection in small ranges. Excellent for “car in the driveway,” “person at the front door,” or “person in the living room.” And with Beecon+, you can define a custom distance as well.

The custom distance thing is really a big deal. I have been working (on Android) to map my home with overlapping beacons to triangulate each room. I am using the Geohopper USB warts, which I just plug into outlets I’ve converted for USB 5V. Ideally, I am shooting for knowing what room a cell phone is in, in every room in my house. Thus far it has been very accurate - but I have not glued the multi-beacon sensing together yet in Tasker.

  1. they don’t toggle for distances other than “out of range.” You could certainly have your lights come up when you to the house and turn off again when you leave. But if you want to use them for Very Close switches, you need one for on and one for off. Or have a metal box to drop over it to force it “out of range.”

I have lobbied Tasker and some of the plugin guys to allow RSSI detection inside BTNear and some of the other Tasker methods. There was a lot of interest, so hopefully you’ll be able to cut out the app and use Tasker directly to do this.

  1. They don’t work great with Android. iOS detects iBeacons at the operating system level, which is both easier on the battery and means the receiving station app can be set to work even when it’s closed. Android’s native functionality requires that the receiving station app be open in order to detect the beacons. I don’t know what other options there are for this, but it’s no coincidence that Beecon is only available for iOS.

Android has really been a chore to get up and running. There are far fewer beacon Apps for Android, and as you mentioned the OS doesn’t handle BLE as well. I believe Android is going to add Bluetooth to its location services – that is the logical thing for them to do – and there are several folks out there mapping BLE infrastructure. Tasker also seems to be interested in using Bluetooth in its “Fused Location” component. The big thing is battery. On Android you can pair to the GeoHopper beacons to cut down battery. I’ve been messing with dead reconning on my Tango tablet to get a high resolution GPS signal at the beacons, and then take a BLE measurement only occasionally. This is really efficient but way far away from production code. I’m pretty confident in 2-3 years this will be the “normal” way micropresence is done though.

  1. unlike a motion sensor, an iBeacon has an actual “exiting” event. So you don’t have to worry about the off event triggering too soon. It’s not based on time, it’s based on you actually moving out of range of one iBeacon (or into range of another).

One of the things I’ve made my ST do is reset the motion to inactive on adjoining sensors. I.e., Hallway motion, then Bedroom motion, immediately set Hallway motion to inactive. This is a simple tweak (requires a custom device) that dramatically improves motion detection reliance. Combined with beacon technology, I think I can get this even better.


  1. every use case depends on you either bringing the phone near the iBeacon or bringing the iBeacon near the detector (which could be a tablet). So it’s definitely not as good as a regular proximity switch for the “Very Close” distance use cases. If you forget your phone in the living room, it won’t trigger the iBeacon in the bedroom. And your phone has to have been set up with the receiving station app first, so it’s unlikely that iBeacons will work well for visitors.

This “split presence” case is nasty and very hard to deal with. I’ve experimented a bit with monitoring Wifi and having APs in every room – this is OK if your guest has your Wifi. I’ve also toyed around with monitoring cell repeaters in the house which works well but you need one per carrier and you have to root the repeater. And those solutions only incrementally help as well. Right now in my house if I can’t match virtual presence to a device / person, I just consider it an intruder :slight_smile: Lots of fun WAF there…

  1. they don’t toggle for distances other than “out of range.” You could certainly have your lights come up when you to the house and turn off again when you leave. But if you want to use them for Very Close switches, you need one for on and one for off. Or have a metal box to drop over it to force it “out of range.”

I think the solution you really want to try here is NFC. Its sort of the “fifth pillar” of location services. It only works at the inch or so level, but its probably more consistent than the metal box. Just put a $1 NFC tag on the table and when you put your phone on that tag, make Tasker turn your switch off. You can also accomplish the inverse pretty well.

  1. If your phone goes completely dead and is then recharged, it will be treated as an Arrived in Range event. Just something to be aware of.

I’ve been working on a SmartApp (Wasp in a box) that creates a virtual presence device per phone per room, and then also creates nested devices (Home -> Upstairs -> Bedroom -> Closet). This generates about a hundred virtual presence sensors in my house (which has its own set of fun challenges) that make it easier for me to associate motion with occupancy and presence. So if I forgot to charge my phone last night, the last known presence sensor was on my bedside table and there was no motion outside of my bedroom, so the presence sensor knows the phone is still there. In the long run, this is almost definitely how microlocation will work at some level – though all the various error correcting is a beast (phone died, motion sensor failed to report, someone exited via a window, etc).

  1. They’re not cheap. If it takes two to handle the on and off conditions, that’s fifty or sixty dollars. Since I’m quadriparetic, there are a lot of use cases where that’s worth it to me, but for many people, the primary justifiable use will be at the geofence border for arriving home/leaving home use cases where you’ll only need one.

The BOM on the Geohopper device, minus the power supply, is about 80 cents. Its very conceivable in the next year or two that these will be so cheap that it will probably be easier to litter your house with beacons than motion sensors. (But ideally both). Being an early adopter is never cheap but its also regularly profitable.

  1. As of this writing, May 2015, they aren’t detectable by smart watches, only phones and tablets. The Beecon+ app does have an apple watch version, but while it lets you launch actions manually, the watch itself doesn’t have the receiving station. So the range is always relative to where the paired phone is, not the watch itself.

From what I can tell my Moto 360 seems to detect the Geohopper beacons just fine. I haven’t messed with any detection stuff yet - but it looks like this is one aspect where Android is further along than iOS. Not sure where Tizen is…


Go into any retail store like a Walmart or Target and turn a Bluetooth scanner on. Its shocking to me how many beacons are already out there.


Yeah, the metal box thing is really a hack for two separate use cases. It won’t be important to most people. Although there’s also a third case with real possibilities.

  1. it’s dog-friendly. I’m quadriparetic, so I really prefer handsfree or touchless switches, as I have very limited hand function. NFC and RFID tags have proven to be too difficult for me, and if I drop one, they’re hard for my dog to pick up. It’s possible to attach the tiny tag to something bigger, but then it can be hard to get positioned perfectly.

The metal box idea is something that’s very easy to set up so my service dog can use. Flip boxes or something like a butter dish with a handle. He can open or close it, no problem. ( Although I would typically wrap the part he grabs in tape as he doesn’t really like the taste of metal.)

Why would this be any better than a conventional switch? Just because it could trigger multiple events or a mode change. And iBeacon plus box is still cheaper than a wired zwave switch. :wink:

But that’s a weird use case, I know. Still, very helpful at my house.

  1. ST, as is often mentioned in the forums, doesn’t have a lot of choices for tactile buttons. There are a few–harmony home hub remote, Minimote, aeon key fob, Securifi key fob, smartenIt 3 toggle switch. All of these except the Securifi cost more than an iBeacon. And the Securifi is so tiny it can be hard to use in the dark. So for the 2 in the morning " turn the dang lights on NOW" case, the metal box might fit. With the range set appropriately, the phone could be in another room. Or you could even use a receiving station app on a tablet control station in that room, maybe a wall mount on the other wall.

For people who care a lot about aesthetics, the box can be ornamental and just look like a decorative piece.

So this use case only exists because there aren’t enough other devices under $35 to do the same thing.

  1. I could also see this as a very practical “panic button” for a child or a physically fragile adult. Open the box, and the lights come on and maybe lights flash in the caregiver’s room or some other signal. Granted, ST itself isn’t reliable enough to use this for life and death situations, but as something more convenient than yelling or phoning, this could be nice. And of course unlike most panic buttons, you can customize it to anything you can control with ST.

And again because you can use the full range, the child doesn’t have to have a smartphone. I do think the ideal for this particular use case is to pair the panic button iBeacon with a receiving station app on a wall mount tablet somewhere. It could manage multiple beacons and it would always be in the right place.

So just some possibilities for the metal box scenario. I agree most people won’t need these, and for many “very close” use cases, NFC will be a better/cheaper choice.


So far I just read thru @JDRoberts detailed explanation, thanks. So this can replace using an iPhone as a presence device, correct? From what I understand so far is it can easily replace using the phone as a presence device and then more, this would be a good alternative for the problematic built in presence sensing.
So I assume I would have one of these iBeacons somewhere in my house and then have the app on my phone which would trigger it when I am close. Can multiple phones work off one Beacon?

Sorry for the stupid questions as this is all new to me :smile:

Yes to all of the above, as long as they’re iPhones and you buy the beecon+ app for $9.

You can install the app on multiple devices under the same license.

Not only that, you can choose to either have the same rules for each device or different ones. So one iBeacon, but it could turn on a different switch for each phone, because it’s the receiving station app that devices what to do when it recognizes the beacon. Not the beacon itself.

Cool. Right now I am just using presence for Arming and Disarming Smart Alarm and turning on a hallway light when all of us are away and at least one of us returns home and it’s night.

Right now I think I only need one to play around with and see how well it works for us. What would you recommend buying?

If you only need one, and you have a place to plug it in, the $25 GeoHopper is the best place to start, I think. You can’t buy just one Estimote, you have to buy a pack of three.

For your particular use case, another option is the Securifi key fob, which costs $15. It’s not automatic detection, but for arming and disarming an alarm mode, it’s a good solution for many people.

Thanks. I was looking for something automatic and was already planning on getting the keyfob as a backup. For $25 I think it’s worth a try to see if it works better for us.
I had bought one of these LightBlue Beans back when they were on kickstarter a while ago but it’s to over my head (still in the box), so the GeoHopper may be my best bet for now.

Thanks for your help.

A friend who is using iBeacons for home automation through Beecon said he saves battery life by putting an iBeacon in his car and using a wall mount tablet as a control panel inside the house near a front window. The tablet detects the car arriving/leaving and that triggers a bunch of stuff.

He also has Beecon installed on his phone, but keeps the regions disabled most of the time in that instance of the app, so it doesn’t drain the phone’s battery.

Then if he’s going out without the car, he has two choices: carry an iBeacon with him or turn on the receiving station on his phone.

For his kids, he just uses an iBeacon the same way people use the ST Presence fob, and puts it on the kid’s backpack. He also put one on the baby stroller. The one tablet at home (which is plugged in) detects everybody’s beacons coming and going.

Interesting ideas. Clearly people are starting to use these a lot of different ways. :blush:



How has this been going for you? I was about to order the $25 GeoHopper since I still have inconsistencies with using our iPhones as presence detectors.


We’re still experimenting with it. Of the three people here, one was having battery issues with their phone running down too fast. Not sure why it was only that one person.

We are now testing having each person carry an I Beacon, and using a tablet at home that is plugged in as detector.

That won’t work with the Geohopper though, only with the battery powered I beacons like the estimates.

So I’m not sure what to tell you. For two out of three of us, the Geo hopper worked great. But for one of us it used too much battery power on the phone.

So this is likely one of those things you just have to try and see how it works in your particular set up.

1 Like

Thats perfect, thanks for letting me know.

1 Like

They killed my battery pretty good, but then again I now have 8 & I was developing against it. So I was pushing the app way more than a “normal person”. I haven’t tested it as a normal user just yet.


@davglass did you pair to the beacon or were you using it differently?

I used geohopper to find it, then paired it to Beacon+, then used it to ping the web endpoint on my SmartApp BeaconBacon.

Just an update:

We’re using the iBeacons more and more as we figure out the best options. But because I’m in a wheelchair, we have done options others won’t, so a lot of people may prefer to wait for wristband or smart watch proximity systems.

  1. permanent stations. We solved the battery drain/phone reboot issues by installing Beecon+ On two tablets that are always plugged in at the house. One in the living room, one in my bedroom. Then attached an Estimote beacon to my wheelchair.

Voila! Some microlocation information at the room level. “JD entered the livingroom.” “JD entered the bedroom.” No false triggers from the dog moving around the house. No network Dropoffs.

However, this still isn’t precise microlocation as I’m not directly tracking exit information ( which iBeacons don’t really do). I’m just capturing entrance events.

  1. I’m just starting to look into the possibilities from the new IFTTT Maker channel. Beecon+ released support for the IFTTT syntax in their most recent release. I know this app costs $9, but I continue to be impressed with both features and reliability.

So my experience with iBeacons has been very good, but also cautionary: phone battery drain is definitely a problem for some people. So I see them as having more use when there’s a fixed receiving station (phone or tablet that’s always plugged in) and it’s the iBeacon that comes and goes with the person. But “all home automation is local,” other people may find the more typical “beacon stays put, phone has the app” approach works fine for their needs. :sunglasses:

1 Like

This is essentially how I envisaged that iBeacon would work, but I didn’t think it would be accurate enough, or rather low power enough not to have two beacons confuse which room you are in.

I personally don’t really care where they in a room just which room they are in.

The iBeacon standard includes three ranges:

immediate: a few inches
Near: a few meters (typical room size)
Far: about 10 meters, can be more

This is all based on the signal strength as evaluated by the receiving station.

Beecon+ also lets you set a custom distance.

The most important thing to note is that the smaller the distance, the more accurate. Immediate range is usually quite precise. I had two ibeacons at opposite ends of an 8’ shelf, used as proximity switches in the immediate range, and both were very reliable. The far range, though, can bounce around quite a bit.

Also important is that because of the variable reporting, Ibeacons do not report an exit event as you move from one range to another. The only exit event is captured when the Beacon signal is lost altogether. So we’re finding that they work very well for capturing arrival events at a specific range, but they’re not quite the same thing as an actual occupancy sensor.

So it all depends on exactly what you’re trying to do. If you do have an I Beacon covering each room, you can certainly write logic and use virtual switches so that a person entering room B by extension means they exited room A and you turn off the presence indicator for room A. But that gets a lot more complicated.

1 Like

Hi all. I just noticed this topic and I am also excited about integrating iBeacons into my smart home. Actually, in trying to find a setup that did what I wanted in my home, I wasn’t happy with what I could do. So, obycode actually has another app, called BeaconThings, waiting for Apple’s approval (in addition to the SmartRules fix). Here’s a sneak peek if any of you are interested - If all goes well, we expect it to get approved later this week. I’ll give away free promo codes to the first 10 interested users to private message me if you wouldn’t mind giving some feedback/suggestions/criticisms after trying it out.


Any updates on BeaconThings? :blush:

This past week, we finally heard back from Apple and have a bit of back and forth about the app. We had to make some adjustments to the metadata and are now waiting for them to re-check it. I’m hoping it will be available this week, but, I could be wrong.

1 Like