There’s new research by ITMO University, Russia on the Implementation of Indoor Positioning Methods: Virtual Hospital Case. The paper describes how positioning can be used to discover typical pathways, queues and bottlenecks in healthcare scenarios. The researchers implemented and compared two ways to mitigate noise in Bluetooth beacon RSSI data.
The probabilistic and neural network methods both use past recorded data to compare with new data. This is known as fingerprinting. The neural network method is less complex when there’s need to scale to locating many objects. The researchers tested the methods at the outpatient department of the cardio medical unit of Almazov National Medical Research Centre.
Comparison of the methods showed they give approximately the same error of between 0.96m and 2.11m. However, the neural network-based approach significantly increased performance.
There’s new research by the Institute of Information Science and Technologies, Pisa, Italy on Detecting Proximity with Bluetooth Low Energy Beacons for Cultural Heritage. The paper starts by describing alternative technologies including Ultra-wideband (UWB), Near Field Communication (NFC) and vision.
The RE.S.I.STO project allows media on the medieval town of Pisa to be accessible via smartphones and tablets. The system is implemented using the React Native Javascript Framework to allow cross-platform aps to be created on iOS and Android.
Beacons are attached to exhibits and the paper compares two proximity detection algorithms, a ‘Distance-based Proximity Technique’ and a ‘Threshold-based Proximity Technique’. The paper describes stress, stability and calibration testing of the system.
RSSI time series of 5 tags
The researchers found a strong variation of RSSI value for different tags that they say is caused by the varying channel (frequency) used by Bluetooth LE as well as environmental issues such as obstacles, fading and signal reflections.
The system was able to successfully detect the correct artwork with an accuracy up 95% using the Distance-based Proximity Technique.
We previously mentioned how cost, battery life and second sourcing are the main advantages of Bluetooth over Ultra-Wideband (UWB). An additional, rarely mentioned, advantage is scalability.
Servers that process Bluetooth or Ultra-Wideband support a particular maximum throughout. The rate at which updates reach systems depends on the number of assets, how often they report and the area covered (number of gateways/locators). Each update needs to be processed and compared with very recent updates from other gateways/locators to determine an asset’s position.
For Bluetooth, updates tend to be of the order of 2 to 10 seconds but in some scenarios can be 30 seconds or more for stock checking where assets rarely move. Motion triggered beacons can be used to provide variable update periods depending on an asset’s movement patterns. This allows Bluetooth to support high 10s of thousands of assets without overloading the server.
For Ultra-Wideband, refresh rates tend to be of the order of hundreds of milliseconds (ms) thus stressing the system with more updates/sec. This is why most Ultra-Wideband systems support of the order of single digit thousands of assets and/or smaller areas. More frequent advertising is also the reason why the tags use a lot of battery power.
How does all this change with the new Bluetooth 5.1 direction finding standard? The standard was published in January 2019 but solutions have been slow to come to the market. The products that have so far appeared all have shortcomings that mean we can’t yet recommend them to our customers. Aside from this, in evaluating these products we are seeing compromises compared to traditional Bluetooth locating using received signal strength (RSSI).
Bluetooth 5.1 direction finding needs more complex hardware that, at least in current implementations, are reporting much more often. The server has to do complex processing to convert phase differences to angles and angles to positions thus supporting fewer updates/sec. Bluetooth direction finding is looking more like UWB in that cost, scalability and battery life are sacrificed for increased accuracy. Direction finding locators are currently x6 to x10 more costly than existing Bluetooth/WiFi gateways. Beacon battery life is reduced due to the more frequent and longer advertising. We are seeing Bluetooth 5.1 direction finding being somewhere between traditional Bluetooth RSSI-based locating and Ultra-Wideband in terms of flexibility vs accuracy.
Despite these intrinsic compromises, Bluetooth direction finding is set to provide strong competition to UWB for high accuracy applications. We are already seeing UWB providers seeking to diversify into Bluetooth to provide lower cost, longer battery life and greater scalability.
The paper starts by describing trilateration and the author voices the opinion that another method, fingerprinting, requires a lot of effort and isn’t feasible for practical implementation.
The new method makes use of the fact that accuracy is usually good when the received signal strength (RSSI) is -70 dBm or better. The use of more beacons and basing calculations on ‘reliable circles’ of higher signal strength, when available, provides for more accuracy.
The data is also filtered using a Kalman filter to reduce signal noise by about 37%.
This article is different because it considers navigation as opposed to just locating. It explains the advantages of Bluetooth LE over WiFi and also compares with RFID:
Trilateration, Received Signal Strength Indicator (RSSI) and Decibel-milliwatts (dBmW) are explained and how these fit into locating position.
The article describes a system created for navigation that uses iBeacon sensor nodes, an Android device and app.
There’s a recent paper by researchers at DeustoTech Institute of Technology, Bilbao, Spain and Department of Engineering for Innovation, University of Salento, Lecce, Italy on Behavior Modeling for a Beacon-Based Indoor Location System.
The research compares two different approaches to track a person indoors using Bluetooth LE technology with a smartphone and a smartwatch used as monitoring devices.
The beacons were iB005N supplied by us and it’s the first time we have been referenced in a research paper.
The research is novel in that it uses AI machine learning to attempt location prediction.
The researchers were able to predict the user’s next location with 67% accuracy.
Location prediction has some interesting and useful applications. For example, you might stop a vulnerable person going outside a defined area or in an industrial setting stop a worker going into a dangerous area.
The paper starts with an overview of indoor positioning techniques including trilateration, fingerprinting, dead reckoning and AI machine learning. It also provides a synposis of different technologies such as RFID, WiFi and Bluetooth.
The paper explains that while fingerprinting is widely used, it faces limitations when used in dynamically changing environments. Fingerprinting requires ongoing maintenance and updating of the reference fingerprinting map that’s manually intensive and time-consuming. Fingerprinting also requires a large number of beacon reference points to perform accurate locating.
The researchers looked into positioning within a two floor (grocery) retail store. Retail stores are of of the more challenging environments as there are shoppers moving about that can affect indoor localisation
Several indoor positioning techniques were considered including fingerprinting and trilateration. The researchers implemented fingerprinting and compared it to seven established classifiers. The random forest algorithm worked the best and inspired the authors to build an ensemble classification filter with lower absolute mean and root mean squared errors.
There’s recent research from Korea on Particle Filtering-Based Indoor Positioning System for Beacon Tag Tracking. The paper looks into how to improve positioning accuracy, reduce system complexity and reduce deployment cost through the use of a Particle Filter-based Indoor Positioning System (PFIPS).
A Kalman Filter is used to preprocess collected Received Signal Strength Indication (RSSI) data followed by a Particle Filter (PF) to approximate the location of a tag which improves the location certainties.
Simulations and experiments showed the system outperformed the legacy indoor positioning systems in terms of location accuracy by 24.1% and achieved median accuracy of 1.16 m.
Areas are differentiated as either being ‘critical’ or ‘common’. For example, in a railway station, critical areas are elevator entrances, boarding gates, toilets and the service centre. Critical and common areas have different positioning needs leading to different sensor deployment densities.
The paper examines the variation of RSSI with distance and develops a critical-grid coverage model. A NSGA-II algorithm is used to optimise the placement of iBeacon nodes.
The results showed that the new placement scheme obtained a lower error and a greater reduction of sensor deployment cost than the uniform deployment scheme. The proposed method reduced the cost of sensor deployment while ensuring the accuracy of indoor positioning for critical areas.
