Creating User Indoor Movement Logs

New research (pdf) looks into the development of an application that tracks user indoor movement logs using Bluetooth beacons. The main focus is on creating a system that is easy to install and use without requiring expertise in beacon installation or positioning analysis. This application is designed for personal home use and simplifies the process by allowing users to install beacons in desired locations, name the spaces and track their movements within the home. The application records users’ movements and the time spent in specific spaces, offering statistical insights such as daily and weekly movement patterns.

The Bluetooth beacons used in this system rely on RSSI (Received Signal Strength Indicator) to estimate the distance between the user’s device and the beacons, with methods like the Kalman filter applied to reduce noise and improve accuracy. To verify its effectiveness, the study conducted experiments comparing manually recorded movement logs with those captured by the application. The results showed an accuracy rate of over 99%, making the system a practical solution for indoor movement tracking in homes, small offices, and other limited spaces.

Key advantages include ease of installation, automatic logging of movement data, and statistical analysis of time spent in different rooms. The application is also suitable for environments like small offices with fewer than 10 employees.

What Bluetooth Systems Can Track Working Using Their Smartphones?

Contrary to popular belief, it’s not possible to directly track smartphones using Bluetooth alone. Both iOS and Android devices have built-in privacy protections and limitations that prevent this kind of tracking.

For iOS devices, Apple has implemented randomised MAC addresses for Bluetooth transmissions. This means that the unique identifier broadcast by an iPhone or iPad changes regularly, making it impossible to consistently track a specific device over time. Android doesn’t continuously send out Bluetooth transmissions.

However, whilst smartphones themselves can’t be directly tracked via Bluetooth, there are systems that can perform location tracking using Bluetooth beacons and gateways. These systems rely on people carrying small Bluetooth beacons, often in the form of keyfobs or badges, which broadcast a unique identifier. Fixed gateway devices are then installed throughout an area to detect these beacons.

When a gateway detects a beacon, it records the beacon’s identifier and signal strength to infer distance, along with a timestamp. By combining data from multiple gateways, the system can estimate the location of the beacon, and by extension the person carrying it, within the covered area. This approach is often used in workplace settings for things like occupancy monitoring or contact tracing.

It’s important to note that these systems require active participation – people must choose to carry the beacon devices. This is quite different from the idea of passively tracking smartphones without user consent.

Some retailers have experimented with using Bluetooth beacons to track customers’ movements within stores. However, this still requires customers to have the store’s app installed and Bluetooth enabled on their phones. These work the other way around by having fixed beacons and the app detecting the beacons. It’s not a covert tracking system, but rather one that customers opt into, often in exchange for discounts or other benefits. It’s less reliable due to the nuances of ensuring the app runs on all phones, at all times.

In summary, whilst it’s not possible to directly track smartphones via Bluetooth due to privacy protections and limitations, there are Bluetooth-based systems that can provide location based services when users actively participate.

Low-Cost AoA Wayfinding

There’s a new paper (pdf) on a low-cost wayfinder system using Bluetooth’s Angle-of-Arrival (AoA) technology. This system is designed to help visually impaired individuals navigate public spaces, such as airports or shopping centres. The innovation lies in moving the antenna array required for angle measurement onto the user’s device, simplifying the beacon infrastructure. Each beacon becomes a low-cost, single-antenna transmitter, significantly reducing the deployment cost compared to traditional indoor positioning systems.

The prototype, built with Bluetooth 5.1 boards and developed using Python, successfully demonstrated accurate angle and distance measurement. The system achieved a 10° angle accuracy within 15 meters and calculated distance using the Received Signal Strength Indicator (RSSI). For visually impaired users, the system could be extended with a voice notification feature. The ultimate goal is to develop the system into a smartphone app.

Future enhancements include addressing front-and-back signal ambiguities by adding orthogonal antennas and extending the system’s range.

Improving RSSI Using Relabelling

Researchers from Japan have a new Relabelling Approach to Signal Patterns for Beacon-based Indoor Localization in Nursing Care Facility. Bluetooth beacons were used in a nursing care facility to enhance the tracking and location estimation of caregivers. These beacons were strategically placed throughout the facility, particularly outside patient rooms and in common areas. The caregivers carried smartphones with a mobile application called FonLog installed, which recorded the Received Signal Strength Indicator (RSSI) readings from the beacons and logged location labels.

The beacons were set to a frequency of 10 Hz with a coverage range of up to five meters. The main challenge addressed in this study was the signal loss and limited data, which affected the accuracy of indoor localisation. To improve the data quality, a relabelling approach was applied. This involved observing the signal patterns in different rooms and using these patterns to augment the training data by relabelling RSSI values from one location as samples for another location with low data samples.

This approach aimed to increase the dataset and improve the model’s accuracy in recognising the caregivers’ locations. By doing so, the accuracy of the indoor localisation model improved, achieving an accuracy of 74%, which was a 5% improvement over the original data. The use of Random Forest for location recognition further enhanced the performance, demonstrating the effectiveness of combining relabelling with machine learning techniques for indoor localisation in a healthcare setting.

Survey of Bluetooth Indoor Localisation

Recent research provides a detailed survey on Bluetooth indoor localisation. The paper underscores the importance of indoor localisation and the unique challenges it presents, such as the inability of GPS to function indoors.

There’s an overview of the types of localisation methods, including triangulation, scene analysis and proximity, as well as the metrics used in these systems. The main localisation techniques discussed are RSSI (Received Signal Strength Indicator), CSI (Channel State Information), fingerprinting and other methods like Angle of Arrival (AoA) and Time of Flight (ToF). RSSI is widely used in Bluetooth localisation but suffers from poorer accuracy due to environmental factors. In contrast, CSI is rarely used due to protocol limitations. Fingerprinting is sometimes employed, involving the pre-collection of measured signal strengths to create a database for location matching.

The survey identifies issues affecting Bluetooth indoor localisation systems, such as accuracy, latency, coverage range, cost and security. Accuracy can be problematic in complex indoor environments, which introduce obstacles and multipath effects that negatively impact signal transmission and reception. The range of coverage is crucial, especially in large indoor spaces where fewer reference nodes are preferred. Cost considerations include both equipment and setup costs, and security issues arise due to the need to protect location data within personal networks.

The study summarises various existing approaches to Bluetooth indoor localisation, categorising them based on their robustness to environmental changes. In discussing RSSI versus fingerprinting, the survey notes that RSSI-based approaches are prevalent due to their simplicity and widespread use. Fingerprinting, on the other hand, involves creating a detailed database of data, which can provide more accurate localisation but requires substantial pre-processing and regular re-calibration to remain effective. Fingerprinting is susceptible to dynamic changes in the environment, making it less competitive in typically fluctuating conditions such as changes in room layout or occupancy.

Indoor Navigation for Environments with Repetitive Structures

New research looks into indoor navigation systems specifically designed for environments with repetitive structures, such as cruise ships, using Bluetooth low-energy (BLE) beacons without relying on GPS. The system incorporates a mobile application that uses these beacons to guide users accurately within buildings. The system optimises navigation through the use of pre-calculated routes, which minimises data storage requirements and enhances the application’s energy efficiency.

It system includes a sophisticated user interface that displays the route and updates navigation in real-time based on user movement and beacon signal reception. The implementation faced several challenges, particularly related to the synchronisation and real-time processing of beacon signals, which were addressed by optimising the beacon scanning process and the communication between system components.

The study lays the groundwork for future exploration and deployment of indoor navigation systems that leverage repetitive architectural features for enhanced navigation efficiency.

Anchor-based Bluetooth Low Energy (BLE) 5.0 Positioning

A recent new paper, BLE-Based Indoor Localization: Analysis of Some Solutions for Performance Improvement, focuses on improving the performance of indoor localisation using an anchor-based system based on Bluetooth Low Energy (BLE) 5.0 technology, specifically employing the Received Signal Strength Indicator (RSSI) for distance estimation. Different solutions to enhance this localisation technology’s performance are explored, with an emphasis on combining various approaches to identify the most effective one. These solutions include different RSSI signal conditioning, anchor–tag distance estimation techniques and methods for estimating the unknown tag position.

An experimental analysis was conducted in a complex indoor environment, marked by the continuous movement of working staff and numerous obstacles. The results showed that the exploitation of multichannel transmission, using RSSI signal aggregation techniques, significantly improved the localisation system’s performance, reducing the positioning error from 1.5 meters to about 1 meter.

Other solutions, such as RSSI signal filtering, distance estimation with an empirical propagation model or Machine Learning (ML), numerical optimisation and ML models for estimating the tag’s unknown position, also impacted performance but to a lesser extent. These solutions resulted in either a decrease or an increase in positioning errors, depending on the specific combination of solutions adopted.

The study’s findings suggest that the use of multichannel transmission and the combination of RSSI signals from different transmission channels are crucial for achieving optimal performance. This approach leverages the full potential of BLE 5.0 technology and is the most significant factor in reducing positioning errors. The paper concludes that the results can guide designers in choosing appropriate solutions based on the desired accuracy of the localisation system. However, it’s noted that the results are specific to the tested conditions and may vary under different operating scenarios.

Indoor Locating Using Beacons in Nursing Care

The new paper Relabeling for Indoor Localization Using Stationary Beacons in Nursing Care Facilities by Christina Garcia and Sozo Inoue from the Graduate School of Life Science and Systems Engineering, Kyushu Institute of Technology, Japan, presents a study on enhancing machine learning for indoor localisation in caregiving, specifically in nursing homes, using Bluetooth Low Energy (BLE) technology.

The study addresses the challenge of limited data available for training machine learning models in indoor localisation, which is critical for monitoring staff-to-patient assistance and managing workload in caregiving environments. The authors propose a data augmentation method that repurposes the Received Signal Strength (RSSI) from various beacons by re-labeling them to locations with fewer data samples, thus resolving data imbalances. This method uses standard deviation and Kullback–Leibler divergence to measure signal patterns and find matching beacons for re-labeling. Two variations of re-labeling are implemented: full and partial matching.

The performance of this method is evaluated using a real-world dataset collected over five days in a nursing care facility equipped with 25 Bluetooth beacons.

Overall, the study highlights the effectiveness of the proposed re-labelling method in enhancing indoor localisation accuracy in nursing care facilities, providing a valuable contribution to the field of caregiving and workload management.

Improved RSSI Indoor Localisation Using AI Algorithms

The article titled Improved RSSI Indoor Localization in IoT Systems with Machine Learning Algorithms by Ruvan Abeysekera and Ruvan Abeysekera focuses on enhancing indoor localisation in Internet of Things (IoT) systems using AI machine learning algorithms. The paper addresses the limitations of GPS in indoor environments and explores the use of Bluetooth low-energy (BLE) nodes and Received Signal Strength Indicator (RSSI) values for more accurate localisation.

GPS is ineffective indoors so the paper emphasises the need for alternative methods for indoor localisation, which is crucial for various applications like smart cities, transportation and emergency services.

The study uses machine learning algorithms to process RSSI data collected from Bluetooth nodes in complex indoor environments. Algorithms like K-Nearest Neighbors (KNN), Support Vector Machine (SVM, and Feed Forward Neural Networks (FFNN) are used, achieving accuracies of approximately 85%, 84%, and 76% respectively.

The RSSI data is also processed using techniques like weighted least-squares method and moving average filters. The paper also discusses the importance of hyperparameter tuning in improving the performance of the machine learning models.

The research claims to provide significant advancement in indoor localisation, highlighting the potential of machine learning in overcoming the limitations of traditional GPS-based systems in indoor environments.

Simple Indoor iBeacon Positioning Method

New research Using iBeacon Components to Design and Fabricate Low-energy and Simple Indoor Positioning Method (PDF) focuses on developing an effective indoor positioning system using iBeacon. The authors propose an enhanced triangulation technique using signal strength signatures for improved indoor positioning precision.

This system integrates a ‘blind’ device and multiple base stations using iBeacon components to form virtual digital electronic fences, effectively receiving signals from moving devices or tags in a targeted area. The proposed method divides the positioning area into rectangular or triangular subareas and establishes a loss value database for improved location estimation.

The system shows high accuracy, with an average error of less than 0.5 m in the worst-case scenario, making it suitable for various environments. The paper covers the architecture of the system, development phases and experimental results demonstrating the system’s effectiveness. The research offers significant insights into low-cost, high-precision indoor positioning methods suitable for diverse applications such as healthcare, smart cities, and industrial settings.