Category: Beacons

Bluetooth Sensor Beacons For Prognostics

If you are considering using Bluetooth sensor beacons for prognostics, you should take a look at the free IEEE paper IoT-Based Prognostics and Systems Health Management for Industrial Applications (pdf).

Prognostics is the determination of health of assets to diagnose anomalous behaviour and predict the remaining useful life. It’s used to:

  • Prevent catastrophic failures
  • Increase asset availability through less downtime and less time wasted through ‘no fault found’ tests
  • Extend maintenance cycles
  • Execute timely repair

The overall aim is to lower lifecycle costs via fewer inspections, repairs and manual inspections. It can be applied to all types of assets across all sectors but is particularly applicable to manufacturing, industry and infrastructure. Infrastructure includes roads and ports as well as utility industries such as water, power and gas.

Prognostics and in-situ testing isn’t new. However, what is new is substantially improved viability and economics. New sensors, such as beacons, are easier to use, can be attached to legacy equipment and have much lower costs. The cost of connectivity and cloud storage is also decreasing. This means more assets can be retro-actively connected and the sharing of data across assets and platforms enables a more complete operating picture. This opens up new business opportunities.

The paper explains the four main types of prognostic management strategies: corrective, fixed-interval preventative, failure-finding, and condition-based maintenance (CBM). It also explains a new fusion approach to prognostics:

The paper gives examples of use of prognostics in the manufacturing, heavy industry, energy generation, transport & logistics, infrastructure assets, automobile, medical, warranty and robotic industries.

It ends with the mention that, in the future, current research and work on energy harvesting will benefit sensors used for prognostics.

More information:

Sensor Beacons
Beacon Proximity and Sensing for the Internet of Things (IoT)
Beacons in Industry and the 4th Industrial Revolution (4IR)
Using Bluetooth Wireless Sensors

Latest Nordic WirelessQ Magazine Available

Beacons are small computers with a complete System on a Chip (SoC). There are four main companies that manufacturer SoCs: TI, Dialog, NXP and Nordic. Nordic is the most popular SoC for use in beacons, mainly because of the lower (tool) license cost and ease for beacon manufacturers developing the software (actually called firmware) that runs in the beacons.

Nordic has a new free Wireless Quarter Magazine that showcases uses of Nordic SoCs in many types of device, not just beacons.

Learn about:

  • Gartner research showing sensor innovation fosters IoT growth
  • Beacons help U.S. shoppers find way
  • Bluetooth LE in Amazon FreeRTOS
  • Bluetooth LE smart textiles on the rise
  • Article combining Bluetooth Low Energy and LPWANs
  • Firmwave’s use of Bluetooth Low Energy beacons to build an inexpensive satellite broadcast system
  • Article on Getting started with Bluetooth mesh

Read more

Resurgence of Beacons in Retail

The demise of Google Nearby prompted some commentators to declare the death of beacons. However, here at BeaconZone we are actually seeing a resurgence of the use of beacons in retail.

Gone are the unsolicited notifications and gone are the ‘get rich quick’ marketers. The scenarios that remain tend to use beacons as an adjunct to something else rather than being the main solution itself. For example, they are used to provide triggering in CloseComm‘s WiFi onboarding app used by Subway, McDonalds, BurgerKing and CircleK and NCR.

Beacons are being rolled out to many food retailers, particularly in the USA. They are also taking new physical forms as witnessed by Mr Beacon:

If you are looking for more innovative uses of beacons in retail, take a look at Alibaba’s Fashion AI concept store as mentioned in the latest Wired (UK):

RFID and Beacons are used to detect items picked up during shopping so that customers can collect what they have looked at, have accessories automatically selected and view what’s in stock. Once they are home, a virtual wardrobe allows customers to buy anything they saw in store.

Beacons can also be used to enable audit compliance. Eric West, Head of Strategy at IMS has a useful free pdf on takeaways from GroceryShop, the retail industry conference. The pdf also mentions the use of beacons in lighting to drive location-based messages and wayfinding. Also:

“Amazon’s 2017 acquisition of WholeFoods was a “tipping point” that ensured all grocery players were speeding up their digital plans.”

Read about Beacons for Marketing

Wiliot To Enable New Beacon Usecases

We mentioned Wiliot last March and since then their R&D team has created early engineering samples that prove it’s possible to create a battery-less Bluetooth LE beacon harvesting energy from radio frequencies (RF).

The Wiliot device looks more like a RFID tag than a traditional beacon in that it’s supplied as a very thin PVC inlay sheet containing the chip and wire antenna together. The thin form factor, no battery and the relatively low cost will allow it to be manufactured into or stuck onto clothing and packaging that will provide for many new usecases.

Producing such a device isn’t easy as it can’t use existing System On a Chip (SoC) devices as produced by Nordic, Dialog and Texas Instruments (TI) because they are too large and use too much power. Wiliot has had to create their own SoC from the ground up, including software tools to develop and program the devices. We have been told it will be a year before Wilot has all the components in place for commercial rollout. Meanwhile, selected organisations can join the Early Advantage Program (EAP). There’s a new a product overview (PDF below) that explains the EAP and the main usecases, connected packaging, connected apparel, logistics and asset tracking:

Wiliot already have Early Advantage Program (EAP) agreements in place with over a dozen brands including top fashion brands, a telco, appliance companies, a furniture brand and packaging companies.

Machine Learning Accountability

AI machine learning is a great partner for sensor beacon data because it allows you to make sense of data that’s often complex and contains noise. Instead of difficult traditional filtering and algorithmic analysis of the data you train a model using existing data. The model is then used to detect, classify and predict. When training the model, machine learning can pick up on nuances of the data that a human programmer wouldn’t see by analysing the data.

One of the problems with the AI machine learning approach is that you use the resultant model but can’t look inside to see how it works. You can’t say why the model has classified something some way or why it has predicted something. This can make it difficult for us humans to trust the output or understand what the model was ‘thinking’ when the classifications or predictions end up being incorrect. It also makes it impossible to provide rationales in situations such as ‘right to know’ legislation or causation auditing.

A new way to solve this problem is use of what are known as counterfactuals. Every model has inputs, in our case sensor beacon data and perhaps additional contextual data. It’s possible to apply different values to inputs to find tipping points in the model. A simple example from acceleration xyz sensor data might be that a ‘falling’ indicator is based on z going over a certain value. Counterfactuals are generic statements that explain not how the model works but how it behaves. Recently, Google announced their What-If tool that can be used to derive such insights from TensorFlow models.

Read about Machine Learning and Beacons

Experiment Before Committing

We see some companies only after they have gone a long way down a particular road only to discover they made a big mistake early on. It might be, for example, they have heavily committed to the wrong beacon, wrong platform or have assumed something on one of the mobile platforms. They didn’t do their research. Often we can help them get on the right track but sometimes not.

At the other end of the research scale we have other companies who ask us “Will beacons work in an xyz environment?” where xyz has ranged, for example, from underground on the tube for the police to inside cars for a car retailer. Taking this further, we also get many, what we call, “armchair entrepreneurs” who want to work everything out before even looking at a beacon.

While we have a lot of expertise and provide advice through consultancy, it’s often the case that there are some aspects that are unknown until things are tried for real in the actual environment. Wireless solutions can be very fickle.

A lot can be learned about beacons, Bluetooth and the environment by buying one inexpensive beacon and trying things out. In the case of software, try implementing a thin slice through the proposed system touching on the perceived risky or unknown areas. Experiment before committing. Don’t go all in buying thousands of beacons and commissioning full custom software until you are confident things will work.

Consider a Feasibility Study

Hybrid Localisation Method

In our previous article iBeacon Microlocation Accuracy, we wrote about ways of using beacon RSSI to determine location. However, what if you were to use and combine beacon RSSI with other ways of locating to create a hybrid method? This is the topic of a new research Unsupervised Indoor Localization Based on Smartphone Sensors, iBeacon and Wi-Fi by Jing Chen Yi Zhang and Wei Xue of Jiangnan University, School of Internet of Thing Engineering, China.

The paper describes UILoc a system combining dead reckoning, iBeacons and Wi-Fi to achieve an average localisation error is about 1.1 m.

The paper also compares the trajectories obtained using different localisation schemes:

Beacon Network Effects

It’s often the case that rolling out beacons for a particular purpose allows them to be put to other unrelated uses. Take, for example, one of our customers, a hospital, who wanted to roll out a network of beacons to allow wayfinding by patients.

Beyond this initial remit, having beacons placed along corridors allowed them to be re-used for asset tracking. Also, the data from both will allow people and asset movement to be monitored to work out choke points as well as ‘wasted’ areas where nothing happens.

If you think wider than your initial requirement it’s often possible to find extra usecases and identify extra stakeholders who might even be able to help cross-fund the initial requirement.

Update: Just after we wrote this, we became aware, via Mr Beacon, that SNCF have a competition to make best use of their beacon network.

Standard vs Proprietary Technology

There’s a thought provoking article, by Lorenzo Amicucci, on the Nordic Semiconductor blog on End-User Factors Impacting Industrial IoT Connectivity. Nordic is the manufacturer of the System-on-a Chip (SoC) in most beacons and Lorenzo is one of their Business Development Managers. While the article talks about Industrial IoT Connectivity and by implication Bluetooth Mesh, the insights are applicable to any project that has to choose between standard or proprietary technology.

The main conclusion is that the best solution from a technical perspective is not always best for the customer. Instead, the best solution should depend on the longer-term business strategy. While a proprietary technology can have the advantage of differentiating your offering it can suffer from future limited supplier availability and possibly shorter lifetime of the technology. Large rollouts:

“…want the confidence that a huge capital spend won’t be wasted on a technology that will be left obsolete in a couple of years.”

More specfically, new and second sourced products from other vendors need to guarantee interoperability for the lifetime of your project.

Read about Generic Beacons

Detecting Temperature With Beacons

Some sensor beacons can be used to monitor temperature. The first thing to consider when comparing temperature beacons is whether they have a dedicated hardware temperature sensor. Some beacons have a temperature sensor inside the main chip (System on a Chip – SoC) that’s less accurate and has less precision. The sensor is mainly there to give an indication of the chip temperature, not the ambient (outside the beacon) temperature. Most beacons only transmit for the order of 1ms every 10 to 5 seconds and enter a very low power state the remainder of the time. This means they not only use low power but don’t significantly heat the SoC. This means the SoC roughly tracks the outside temperature.

In our sensor beacon listings, when we say a beacon has a temperature sensing it has a separate hardware sensor, usually the Sensirion SHT20, providing more accuracy and precision than the sensor in a SoC. Some of our beacons, such as the Minew i3 and i7 have an internal SoC temperature sensor that’s readable but we don’t classify that as a sensor beacon.

The next thing to consider is the casing. In order to quickly track ambient temperature, the casing needs to be open somewhere and usually have a hole. Beacons that say they are waterproof and have temperature sensing won’t track ambient temperature well.

We have had customers use temperature sensing beacons in scientific situations and where they need to periodically calibrate sensing equipment. How do you calibrate temperature sensor beacons? The SHT20 is has a long term drift of only <0.04 deg C/year (the humidity reading vaies difts by <0.5%RH/year) so it doesn’t need calibration for most situations. However, if you need better than this, or check calibration, you will need to periodically calibrate in the software of the device (usually an app) that receives the beacon sensor data.