Indoor location and micro-location systems are complicated by the lack of available GPS signals. This gap is being filled by Bluetooth and WiFi, but these systems have dificulty maintaining accuracy when the user is moving. This paper presents the results of an exploration of new features of Bluetooth 5.2, namely, Angle of Arrival and Angle of Departure using a Texas Instrument development board and Antenna Array. The research results are: 1) a novel prediction system for indoor positioning and navigation that performs at an accuracy of 0.23m (static), and 0.30m (moving); 2) a comparison and performance analysis of micro-location algorithms; and 3) an architectural model by which other researchers can extend our work on indoor positioning and navigation.
Global Positioning Systems (GPS) are an integral part of our day to day lives. GPS signals assist
us with road transportation, aviation, shipping, rail transportation, science, security, mapping
and several other applications [
This paper is structured as follows: section 2 provides a literature review of related work in this field, section 3 presents our methodology on the design, development and evaluation of our micro-location system, section 4 presents the findings, section 5 provides a discussion, and section 6 provides a conclusion and suggestions for future work.
Indoor positioning systems have gained considerable attention in recent times. The application of indoor positioning can be adapted to indoor navigation and tracking of movement with suficient amount of well placed Bluetooth Low Energy (BLE) beacons. A suitable dynamic positioning system to meet the desired goal of sub-2m accuracy in indoor environments would need to be able identify location of static BLE beacons and the dynamic location of the user (or asset) with respect to the beacons. There are numerous applications where this degree of accuracy is needed. For example, hospitals need to track the location of wheelchairs, patients and many pieces of medical equipment; warehouses need to track the movement of products and supplies; large educational institutes can benefit from students/teachers finding classes easily; geo-fencing around construction sites so workers do not go to certain areas without proper protection [12, 7, 13, 14]. There is a need across these sectors for a solution that provides a level of accuracy that is much higher than current solutions provide [4, 5]. Various studies have been conducted using technologies such as Wireless Local Area Network (WLAN) [12], BLE [15, 16], Radio Frequency Identification (RFID) [ 17], Ultrasonic waves, and ZigBee [18]. However, WLAN [19] and BLE [20] are the most popular due to the ease of deployment, cost, availability of technology on various devices. This section presents a comparison and analysis of the state-of-the-art in micro-location using the following techniques: Wi-Fi Based Indoor location, RFID Based Indoor location, and BLE beacons.
Wi-Fi solutions use existing wireless networks and infrastructure within a facility to determine the location of an asset or person. Since most businesses already have this type of infrastructure, it is the easiest and most cost efective approach to deploy [ 21]. However, there are several issues with using Wi-Fi based systems such as the distance between existing wireless access points may be too large, the inability to move access points easily to improve location accuracy and the cost of enterprise level wireless access points [4, 8]. A recent study [22] showed an accuracy of 1.42m in a 8 x 8 x 3 room with 4 Access Points and 9 testing locations. However, the amount of data collection required was around 10,000 data points [22]. Unless the Wi-Fi access points are already deployed: the start up cost of Wi-Fi access points as location tracking system is considerably more than Bluetooth beacons (3-4 times more). For instance, the Gimbal beacons are $15-$20 USD [13].
RFID solutions have two separate strategies for solving the problem of micro-location: active RFID and passive RFID [12]. Active RFID is an electronic device that either broadcasts or reads RFID signals, for example RFID reader/broadcast chips in modern smartphones [9, 23]. Passive RFIDs are commonly inert chips that use the build-up of electrical signal from an active RFID reader. Once an acceptable charge is established, the passive RFID emits a short broadcast. RFID stickers and swipe cards are a common example of this type of RFID technology [17]. Due to its restricted broadcast distance [24, 17], RFID is a poor choice of technology for micro-location research.
BLE became part of the Bluetooth standard in 2010 with Bluetooth Core Specification 5.0 active since 2016 [25]. BLE enables a device to use Bluetooth networking at lower energy levels in an efort to reduce battery consumption [ 11]. A BLE beacon is a small device that emits a Bluetooth Low Energy signal. This difers from traditional Bluetooth in that the signal and device are essentially read only. There is no connection to these beacons as their purpose is exclusively to broadcast information [25]. The beacon sends out a packet, which is unique to each manufacturer. It contains specific information for the receiving device as shown in Fig. 1. Beacons can be incorporated into 1micro-location systems using the beacon’s configuration information and the RSSI. The RSSI is useful in approximating the distance from the transmitting device. For example, in the iBeacon protocol, this framework provides some interpretive measurements of distance using RSSI, and using three zones (i.e., Immediate, Near, and Far) [23]. Intended to provide an approximate guide for proximity, this approach is far from an accurate measurement and thus unsuitable for micro-location [23, 26]. The concept of indoor navigation presents some unique concepts and challenges. As the user is on the move and location is dynamically changing, it is dificult to accurately calculate the user’s position. An approach to overcoming these challenges is to use algorithms that can quickly identify nearby beacons and use various formulas to determine the location. A recent study [27] explored various approaches to indoor navigation using Wi-FI and BLE using triliteration and a path-loss model in an ofice setting. The accuracy reported for BLE was around 6m [27].
At the onset of this research, we recognized that there are only a few hardware companies that build iBeacon compliant beacons. Our aim was to select the most appropriate product for our purposes. Using the following criteria: 1) accuracy, 2) availability, 3) price, 4) well-designed and supported Software Development Kit (SDK) 5) secure use and communication, 6) ease of use and configuration, 7) data analytics, and 8) AoA and AoD. We reviewed the following beacon 1Micro-location is the process of pinpointing a person’s placement to within a few inches or feet using various technologies. While GPS can only determine geo-location while outdoors, micro-location technology can determine location more precisely, both indoors and out. products: Onyx Beacons [28], Bleu Station Beacon Series 100 [29], Estimote [30], Verve [31], Gimbal [13], Kontakt.io [32], and Texas Instrument’s LAUNCHXL-CC26X2R1 development board. The results are presented in Table 1. Regarding micro-location and the accuracy of beacons, equation 1 shows the relationship between RSSI and distance [33]. We used this equation to evaluate the accuracy of diferent beacon products when compared to ground truth (exact distance measurement from smartphone to beacon), where represents a path-loss exponent that varies in value depending on the environment, is the distance between the user and the beacon, 0 is the reference distance which is 1 meter in our case, and is the average RSSI value at 0.
= − 10 log10 ︁( 0 )︁ + (1) Bluetooth 5.1 introduced two new features called Angle of Arrival (AoA) and Angle of Departure (AoD) that allow for highly accurate positioning of Bluetooth devices [25]. AoA and AoD angles can be calculated from signal transmissions that land onto the receiving arrays (AoA), and for AoD, the departure angle emanating from the transmitter device. Bluetooth specification 5.2 and 5.3 provide no additional improvement with respect to AoA and AoD [25]. AoA allows a receiving device to determine the direction from which a signal is coming, while AoD allows a transmitting device to determine the direction to which a signal should be sent. This is achieved by using antenna arrays and measuring the phase diference between signals received by diferent antennas. By using AoA and AoD, Bluetooth devices can determine their precise location in three-dimensional space, enabling a wide range of applications such as indoor navigation, asset tracking, and proximity-based services [25]. The accuracy of this technology depends on the number of antennas used and the signal processing algorithms employed, but it has the potential to be substantially more accurate than other positioning technologies [25].
The primary goal of this project was to create an indoor navigation system that predicts the user’s location as accurately as possible if the user is standing or while the user is in motion. The secondary goal was to create a rapid deployment strategy that could be used in a variety of environments and scenarios without the need to have training data (i.e., as is the case of ifngerprinting solutions). Our aim was to create a system that uses BLE beacons that consistently and accurately pinpoints the user’s location in the sub 2m range while the user is moving. To our knowledge, there are no such systems that use BLE beacons that report this degree of accuracy. This section presents the methodology by which we created and evaluated our indoor navigation system: 3.1.Experiment setup: equipment and environment configuration; 3.2. Data collection strategy; 3.3. Static test; and 3.4. Moving test.
We used hardware development kits designed by Texas Instruments to test AoA and AoD, and develop our system. Regarding the microcontroller, we used the LAUNCHXL-CC26X2R1 board. This microcontroller features an ARM Cortex-M4F processor, which runs at a speed of 48MHz, along with a wide range of peripherals such as a 128x128 LCD display, two user buttons, and an RGB LED. It also includes an integrated 2.4GHz radio, which supports Bluetooth Low Energy, Zigbee, and Thread protocols. Fig. 2 presents the LAUNCHXL-CC26X2R1 development board. For the antenna, we used BOOSTXL-AOA (Fig. 3). It is a direction finding antenna array board that allows users to determine the AoA of RF signals in the 2.4GHz band. The board features four antennas spaced at a known distance, which are used to determine the direction of incoming signals. In our experiments, we designated the BLE transmitter as a slave device and the antenna array as a passive device. For both slave and passive device, the TI LAUNCHXL development board was used with the exception of passive devices having the BOOSTXL board attached. The setup is presented in Fig. 4.
Initially, we collected data on angle testing to figure out the best way to utilize the BOOSTXLAOA antenna array. The board has two antenna arrays and when using it in single antenna array it provides a -45°to +45°angle. When using two antenna arrays at the same time the angle of arrival calculation goes from 0°to 100°on one side and 0°to -100°on the other side. This can be seen on the Fig. 3
The slave device was set at 0°and 45°angles for various distances and error in angle was measured. For each experiment data was collected for 10 minutes and then averaged out. The experiment was done in an ofice room with Wi-Fi signals and other electronic components like computers, monitors and reflective surfaces to simulate real world scenarios. The test was performed in where the Bluetooth slave and passives devices will always be in a Line of Sight (LOS) scenario. Through the experiment we concluded that using 1 antenna set up yields the best result. The results can be seen from Table 2 where is the standard deviation of the error in angle received. Since the 2 antenna array results in worse results, for all the other experiments moving forward: the second array was disabled using the firmware options.
From the TI passive device, we can get the value of RSSI. A Kalman filter was used to enhance the robustness, accuracy, and reliability of BLE signals which are often noisy. The RSSI value can be used to determine the distance from the transmitter. There are a few techniques to calculate RSSI with respect to calculating the distance. One of the ways to calculate distance is the outdoor propagation model using free space path loss (FSPL) formula in equation 2.
() = 20 log10 () + 20 log10 () + 92.45
However, for indoor transmission, it is dificult to use a specific model as the propagation
varies drastically based on various factors. These include type of indoor environment, position
of the transmitters within this environment, distance from walls, height of the transmitter
compared to the ground and location of obstacles such as furniture [
= +
The model uses a power regression analysis to calculate the constant values of , and . With a base RSSI calculated at 1m distance, a ratio is calculated for RSSI values at various distances. Using the RSSI, Ratio, and Actual Distance we can calculate the Predicted Distance. From our testing we were able to create the following model for 1 passive BLE device (equation 4). 0.59 * ︂( )︂ 6.7 − 48 − 0.14 This formula provided the most accurate distance based on RSSI for our ofice indoor environment. (2) (3) (4)
In order to collect accurate angle of arrival data, we created a testing grid of 9 points across a 2m x 2m and 4m x 4m area. Fig. 4. illustrates a testing scenario where a 2m x 2m grid is used to collect AoA data using one antenna (passive device) and one Bluetooth receiver (slave device).
An optimum height of 1m from floor was used for all data collection processes. Data was collected for 5 minutes on each testing location. The results were then averaged over all 9 testing points for better reflection of accuracy. The data that was received form the Bluetooth Receiver (slave) is in the form of AoA and RSSI. The RSSI value gives us a radius of possible locations and the AoA value solidifies the position on that radius. The error in distance is calculated from ifnding the diference in positioning of actual testing point and measured location. We created a Web application that takes these measurements from the Bluetooth Receiver and plots them on a graph to show the real time location of the Bluetooth receiver. Fig. 5 shows the output when a receiver was placed 0.5m away at an angle of 0°.
For the static test, the slave device was placed at various locations of our testing grid. The device was not moved for the duration of data collection. The initial testing was done with one slave device and one passive device. This gave us a baseline accuracy for AoA tests. Using the libraries available for the TI board we created a web application that shows the RSSI values in real time. We superimposed the figures of passive devices and created a grid with accurate distances in order to visualize location and signal directions that were received on the passive devices. Fig. 5 shows a screenshot from our web app for a static test at 0.5m distance.
To calculate the distance between the slave and Two, Three, or Four passives, the setup relies on using two passives’ Points of Intersection. If there is no direct point of intersection, then the application forces it to intersect. If 2 lines do not intersect, since 1 line may be shorter than the other, then it will force the shorter line to intersect with the other line. Fig 6 illustrates the process. In each of the testing grids, we used a diferent number of passive devices to calculate optimum number of Bluetooth sensors and their location.
The moving test involved the same setup, with the exception of the slave device being moved at a constant pace for the entire test. This can be visualized from the following Video on YouTube. We created a mini-robot using Lego Mindstorms EV3 to move at a constant speed through our testing locations. Data was collected for two scenarios: a) the robot stopped at the testing locations for two seconds and b) the robot moved from endpoint to endpoint without stopping. Data was collected every 0.25s and averaged out for better accuracy.
Table 3 presents the results of our static and moving test results. The results are presented in terms of accuracy in metres for 2x2m and 4x4m grid environments with percentages representing the amount of change by adding more passives. The moving test results naturally are worse than static tests as signals experience Doppler efect, multi-path interference, reflections and scatterings [34]. The most obvious observation is that as we increase the number of passive devices, there is a diminishing return. Three passive devices produced the best results in our environment. The accuracy increased when increasing passive devices from 1 to 2 by 65%, whereas moving from 2 passive to 3 passive produced an increase of only 6.75% for the 2x2m grid. Similar results may be seen for the moving tests, however, the gain is not as large as the static tests. When the slave device was stopping at the testing locations, the greatest improvement in accuracy was the 3 passives scenario (i.e., an improvement of 32%). For moving tests, when the slave device did not stop and continued at a constant speed, the improvement was the most at 2 passives (60% at 2x2m grid). The results also show a similar pattern for the 4x4m grid configuration. The only exception was for the 2x2m grid and moving test, where there was no improvement for moving accuracy when adding an additional passive. Furthermore, there was no change for adding a 4ℎ passive device.
The main aim of this research was to create an indoor localization and navigation system that could be easily deployed with commonly available beacons and smartphones/tablets, while achieving a 2m level of accuracy while the user is in motion. Our system enables eficient real-time tracking of users or assets in indoor environments with a level of accuracy that is an improvement on previous work in this area.
The benefits of AoA/AoD have been well documented. There are improved navigation and augmented reality benefits [ 35], along with enhanced security [36], improved network eficiency and better user experience . However, all these studies, including ours, do not use a mobile device that can use AoA/AoD. At the time of this publication, the authors could not find any mobile devices that has the AoA/AoD antenna built in. Most current day smartphone have access to BLE 5.2 technology, but not the antenna array required to calculate AoA/AoD signals. Our research shows the benefit of utilizing the BLE antenna arrays for indoor navigation along with other aforementioned benefits.
the research conducted demonstrates a framework for indoor localization using AoA and AoD. This can be scaled up for larger areas where the grid is limited to 4m x 4m area. Our test area was a normal ofice indoor environment. Future work could include diferent indoor environments like malls, warehouses and areas with obstacles and reflective surfaces. The idea would be to create a mathematical model that can be easily scaled to cater for various environments. This can help in calculating how many passive devices would be necessary along with their positions.
In this work, we used a Texas Instrument development board and Antenna Array to conduct a thorough set of experiments exploiting AoA and AoD in BLE beacons. We created an indoor navigation system that improves the beacon’s RSSI signals and uses various algorithms to provide high accuracy location predictions in indoor environments. Our work can be used in a variety of applications in indoor location and asset tracking. The main contributions of this work are: 1. An indoor positioning system that performs at an accuracy of 0.23m (when the asset is static) and 0.30m (when the asset is in motion), 2. A generalized indoor micro-location system that can be easily and rapidly deployed to new environments (within hours), 3. A comparison and performance analysis of BLE passive devices and how many passive is required depending on the scenario of diferent area and requirement of static or moving objects. 4. An architectural model by which other researcher can extend our work on indoor positioning and navigation.
In conclusion, we created a real-time context aware solution using BLE beacons with AoA and AoD for indoor environments that can track the user or asset while in motion.
In the spirit of furthering science, the source code for the apps, architectural designs, algorithms and datasets will be openly available on the publisher’s website. We hope this will encourage others to extend our work.
We wish to acknowledge the contributions of our research students, Stefano Gregor Unlayao and continued support from our industry partner SOTI Inc. [4] A. Booranawong, K. Sengchuai, N. Jindapetch, Implementation and test of an rssi-based indoor target localization system: Human movement efects on the accuracy, Measurement 133 (2019) 370–382. [5] E. R. Sykes, A context-aware system using mobile applications and beacons for onpremise security environments, Journal of Ambient Intelligence and Humanized Computing 11 (2020) 5487–5511. URL: https://doi.org/10.1007/s12652-020-01906-2. doi:10.1007/ s12652-020-01906-2. [6] A. N. Raghavan, H. Ananthapadmanaban, M. S. Sivamurugan, B. Ravindran, Accurate mobile robot localization in indoor environments using bluetooth, in: 2010 IEEE international conference on robotics and automation, IEEE, 2010. [7] H. O. Maycotte, Beacon technology: The where, what, who, how and why, 2015. URL: https://www.forbes.com/sites/homaycotte/2015/09/01/ beacon-technology-the-what-who-how-why-and-where/#4baf7381aaf8. [8] E. R. Sykes, S. Pentland, S. Nardi, Context-aware mobile apps using ibeacons: Towards smarter interactions, in: Proceedings of the 25th Annual International Conference on Computer Science and Software Engineering, ACM, USA, 2015, p. 120–129. [9] X. Jiang, Y. Chen, J. Liu, Y. Gu, L. Hu, Fselm: fusion semi-supervised extreme learning machine for indoor localization with wi-fi and bluetooth fingerprints, Soft Computing 22 (2019) 3621–3635. [10] T. Kawai, K. Matsui, Y. Honda, G. Villarubia, J. M. C. Rodriguez, Preliminary study for improving accuracy on indoor positioning method using compass and walking detect., in: International Symposium on Distributed Computing and Artificial Intelligence, 2018. [11] M. Wooley, Bluetooth Core Specification Version 5.1 Feature Overview, report, Berkley, 2020. URL: https://www.bluetooth.com/bluetooth-resources/ bluetooth-core-specification-v5-1-feature-overview/. [12] O. Belmonte-Fernandez, A. Puertas-Cabedo, J. Torres-Sospedra, R. Montoliu-Colas, S. Trilles-Oliver, An indoor positioning system based on wearables for ambient-assisted living, Sensors 17 (2017). [13] Gimbal, Gimbal proximity beacon series 21, 2021. URL: https://docs.gimbal.com/index.html. [14] D. Giovanni, Luca, A. Machì, Towards accurate indoor localization using ibeacons, fingerprinting and particle filtering, in: International Conference on Indoor Positioning and Indoor Navigation, 2016. [15] B. Benaissa, F. Hendrichovsky, K. Yishida, M. Koppen, P. Sincak, Phone application for indoor localization based on ble signal fingerprint, in: 2018 9th IFIP International Conference on New Technologies, Mobility and Security, 2018. [16] A. De Blas, D. López-de Ipiña, Improving trilateration for indoors localization using ble beacons, in: 2nd International Multidisciplinary Conference on Computer and Energy Science, Springer, 2017. [17] E. Hatem, B. El-Hassan, J.-M. Laheurte, S. Abou-Chakra, E. Colin, C. Marechal, Study the estimated distance error in indoor localization using uhf-rfid, 2018 IEEE Middle East and North Africa Communications Conference (2018). [18] D. Konings, N. Faulkner, F. Alam, F. Noble, E. M. Lai, The efects of interference on the rssi values of a zigbee based indoor localization system, in: 24th International Conference on Mechatronics and Machine Vision, IEEE, 2017.