Bluetooth Power Consumption in Phones

A growing use of beacons is for continually monitoring using sensor beacons. However, one concern is how continually reading a Bluetooth sensor beacon might affect phone battery life.

There’s a recent research paper by Kleomenis Katevas, Hamed Haddadi and Laurissa Tokarchuk of Queen Mary University of London, UK on Power use : SensingKit: Evaluating the Sensor Power Consumption in iOS devices. It looks into Bluetooth beacon (phone) power use and, as a baseline, compares this to phone battery power use by sensors in the phone. They evaluated beacon broadcasting and scanning modes separately and together.

Very few scenarios use the phone to broadcast so the pink dotted line probably has less relevance. Looking at the scanning test it can be seen that it consumes power of the same order of magnitude as other sensors in the phone itself. Bluetooth LE scanning isn’t especially power hungry. Nevertheless, the 25hrs hours operation time on battery might be a limitation for most sensing scenarios.

An omission in these tests is that they only considered scanning rather than connecting. Some beacons need to be connected to via Bluetooth GATT, to obtain sensor data, that uses more phone battery power because it requires the phone to transmit to the beacon instead of just listening to the advertising scan data.

In practice, sensing projects are often better served by using a WiFi gateway or a phone/tablet permanently plugged in rather than a user’s battery-powered phone. This allows the device receiving sensor data to be mains powered removing concerns regarding device power use.

Beacons in Outdoor Learning Spaces

If you are working in learning, you might want to take a look at a new research paper Designing Outdoor Learning Spaces With iBeacons: Combining Place-Based Learning With the Internet of Learning Things (pdf). It takes a long time to download because it’s part of a compendium of research from the 12th International Conference of the Learning sciences.

The paper discusses place-based science learning in the context of an arboretum. It describes the use of beacons for informal learning, place-based learning, and context-sensitive educational technologies. Content was delivered in the arboretum with question prompts and activities so as to encourage deeper learning of plants

The paper says it has:

“insights into designing for learner-centered mobile computing that moves beyond presenting just-in-time information to creating digital-physical spaces where learners engage each other and natural objects to support their interests in science.”

“Beyond presenting just-in-time information” is an interesting aspect in that it can be just as applicable to other beacon usecases.

System-based Beacon Detection

The majority of beacon-based solutions are app-based and trigger information to be displayed to the user in response to being near specific beacons. If you read many platform provider sites you might think that’s all beacons can do. However, beacons are a technology and not solution. Beacons provide for many types of solution.

Another type of solution is the accounting for things (with beacons attached) within a larger system. Examples include class registration, stock checking, asset tracking, security and lone worker positioning. In these cases the thing that detects beacons can be can be an app or hardware.

The app can be relatively simple and scan for particular beacons and save information to a file and/or send them on to server. We recently implemented such a system for Malvern Instruments, with custom pre-configured beacons, that also allows search for particular ‘lost’ beacons:

malvernsolution

In the cases where the beacon detection doesn’t or shouldn’t move around, it’s possible to use gateways to forward on detected beacon data to a server.

igs01_smaller

IGS01 Wifi Beacon Gateway

Several of our clients are using this type of architecture to provide for automatic human registration/rollcall type solutions.

We believe even greater opportunities exist for IoT scenarios where sensor data in beacon advertising can be automatically forwarded on to servers.

The Complete Beacon Industry Report

The Proximity Studio has a useful new document (pdf) The Complete Beacon Industry Report with industry insights and usecases. It covers opportunities in manufacturing, retail, facilities management, logistics and healthcare.

It also has interviews with Szymon Niemczura of Kontakt.io and Steve Statler author of the book The Hitchhiker’s Guide to The Beacosystem.

beaconsinmanufacturing

Beacons in Manufacturing

Learnings from Using iBeacons in Wales’ Oldest Gallery

There’s a useful article on the Nesta site on Using Proximity Technology to Enhance the Gallery Experience.

Oriel Plas Glyn-y-Weddw in Llanbedrog on the Llyn Peninsula is Wales’ oldest art gallery. They created a mobile app that uses iBeacons to deliver content to gallery visitors.

They have some insights:

  • They found that audio-only content was best so as not to distract from the art itself.
  • Users were most interested in content presented by the artists themselves rather than other commentators.
  • Positioning the beacons was important. Planning and positioning of beacons was vital in ensuring a glitch-free experience.

Our experience of using beacons in art galleries shows that, as with Oriel Plas Glyn-y-Weddw findings, most problems occur when beacon transmissions overlap. You have to fine tune beacon power and/or trigger on specific ranges in order to prevent false triggering or ‘bouncing’ between exhibits when the user hasn’t even moved. Apps can also be set to ignore multiple triggers that happen within a very short time.

Beacons For Bees

There’s an interesting new project on GitHub that uses Eddystone-URL beacons to tag wild and domestic beehives.

“There are many reasons to geo-tag wild and domestic Beehives, one is simply to raise awareness of HoneyBee Colony Collapse Disorder (CCD), and the state of health of local Beehives; another would be to alert those that might be susceptible to anaphylactic shock that they should be mindful of their surroundings. (i.e. Don’t climb that lovely tree with the huge wild Beehive in it…)”

It’s questionable whether Eddystone-URL is the best solution in this particular scenario. Eddystone-URL will only show up when users are interacting with their devices (when the screen is on). People avoiding beehives due to possible anaphylactic shock would want to be alerted even when not using their phones. This requires use of an app and iBeacon if background notification is required on iOS.

Nevertheless, Eddystone-URL does provide an inexpensive, easy to create solution for educational and awareness (PR) purposes.

Crowd Analysis Using Beacons

With so many uses of beacons centred around notifications to users, it’s interesting to see Queen Mary University of London doing something different. Research by Kleomenis Katevas, Laurissa Tokarchuk, Hamed Haddadi and Richard G. Clegg of the Department of Computing of Imperial College looks into detecting group (crowd) formations using iBeacon (pdf).

They used beacon RSSI and phone motion together with algorithms based on graph theory to predict interactions inside the crowd. They verified their finding using using video footage as ground truth.

distanceestimationmodels

The paper has some particularly interesting observations from testing RSSI in an EMC screened anechoic chamber and also has some information on distance estimation models.

Using Beacons with Gamification

We have previously written about trying to achieve ‘want-in’ rather ‘opt-in’ for beacon apps. One way to achieve this is to use a technique called gamification that is often used in non-beacon apps to encourage greater engagement and retention:

“The application of typical elements of game playing (e.g. point scoring, competition with others, rules of play) to other areas of activity, typically as an online marketing technique to encourage engagement with a product or service.”

There’s a recent paper by the University of Cambridge Museums consortium, a group of eight museums within the University, working in partnership also with the University’s Botanic Garden and other University collections written by Tommy Nilsson, Alan F. Blackwell, Carl Hogsden and David Scruton. The paper, ‘Ghosts! A Location-Based Bluetooth LE Mobile Game for Museum Exploration’ (pdf) examines the challenges and opportunities introduced by the introduction of a beacons into museum scenarios.

They found that the nature of the museum interior with large shelves and artefacts scattered throughout the space, contributed to unpredictable signal coverage. Instead of trying to implement traditional location finding, they instead used gamification to have ‘ghosts’ popping up on the screen of the mobile device explaining to the player that they are lost and need help finding their way back to their home artefacts. The unpredictable signal coverage became part of the challenge and game. They describe the result as a:

“Museum guide masking itself as a hide and seek game”

One of their conclusions is that:

“BLE can thus play a significant role in advancing the internet of things from a mere futuristic vision to a mainstream level of use”

Transport Information System Learnings

There’s a recent article on ZDNet on Smartphones, Bluetooth beacons: The pairing that could help the blind catch the right bus. After weeks of testing, the city of Strasbourg in France is now ready to rollout a smartphone-based transport information system. 1,400 beacons at libraries, tourist spots and bus stops allow users to interact with city locations on their smartphones.

Two interesting issues with the initial trial were:

  • Slow detection time meant users at bus stops were being notified too late
  • Placing beacons behind the windscreens of busses, the 80m range wasn’t sufficient

While the article doesn’t explain how the slow detection time was solved we suspect it had something to do with the app having to connect to get information. This information might have been bus information or something to do with the beacon platform. Caching information or turning off advanced functionality that required the app to connect will have solved the problem. Alternatively, it could have been that the time between beacon advertising was too long preventing it from being picked up quickly by the app.

The 80m range was solved by configuring the beacons to be more powerful and transmit to 160 meters. Unless the beacon was already transmitting at less than normal (0dBm) power it wouldn’t have been possible to double the range by re-configuring the power. Most beacons go up to +4dBm which, while providing over double the power, wouldn’t have doubled the range. It’s more likely they had to use a different beacon with integrated power amplifier such as those in our Ultra Long Range.