IoT Localization Systems Based on a Single AoA Estimation Unit: An Application for Smart Pedestrian Crossing
Noori BniLam, Dennis Joosens, Rafael Berkvens, University of Antwerp - imec, IDLab - Faculty of Applied Engineering, Belgium; Jan Steckel, University of Antwerp, Cosys-lab Research Group, Antwerp, Belgium and Flanders Make Strategic Research Center, Belgium; Maarten Weyn, University of Antwerp - imec, IDLab - Faculty of Applied Engineering, Belgium
Location: Atrium Ballroom
Alternate Number 3
Over the past years, communication systems have been under continuous developments. Most of these developments were concern with increasing the throughput of the communication system and decreasing the latency. Nowadays, the industry demand has been shifted towards the ‘smart’ systems idea. Accordingly, the requirement for smart devices (i.e. devices can connect to the internet and can perform tasks without any human interaction) starts to increase. This demand has been the key factor of the birth of the Internet of Things (IoT) technology. IoT technology allows any device to be connected to the internet via a cost effective and low power transceiver. Shortly after the introduction of the IoT technology, the concept of smart cities starts to verbalize. Consequently, many places all over the world started to deploy this concept. For instance, in Antwerp/Belgium, a part of the city has been dedicated for being a smart zone (where all the available connectivity technologies are welcome to be deployed).
In this paper we present the concept of smart pedestrian crossing as an important part of the smart city concept. Providing a safe and a smooth flow of people over a pedestrian crossing can be a very challenging task. Depending on the age and the health, people’s speed varies in walking over the pedestrian crossing. Therefore, in many occasions, the traffic lights turn green for the cars while people are still walking over the pedestrian crossing. In these occasions, the traffic flow will be disturbed, and the safety of the people will be jeopardized. Therefore, in such an occasion, the traffic can be handled smartly, by keeping the traffic lights red for the cars allowing the safe crossing for the pedestrian, meanwhile other traffic lights can be allowed to keep the green lights for longer period. Therefore, by adopting this approach, a better traffic flow in the city can be achieved.
Solutions based on cameras can be adopted to monitor the pedestrian crossing. Such solutions are not preferred due to privacy issues and the dependency of cameras on clear visual sight of the pedestrian crossing (which is not always guarantied). Furthermore, the deployment of cameras with processing unit in every pedestrian crossing is a high cost solution. Here, we propose a solution based on IoT technology combined with Angle of Arrival (AoA) estimation technique to localize people that are walking over a pedestrian crossing.
AoA-based localization system depends on the angular information of the received signal to estimate the location of the transmitter. AoA-based localization systems require multiple array antenna systems to be distributed in the vicinity of the transmitting device. Every array antenna system can (based on an AoA estimation technique) provide the direction of the received signals with respect to the array antenna system. Combining the direction information, either by a triangulation process  or probabilistic models , can provide a location estimate of the transmitting device.
In this paper, a new AoA-based localization solution based on a single AoA estimation unit (that is suitable for the smart pedestrian crossing application) is introduced. The proposed solution entails that the pedestrian should have an IoT transceiver that gets activated as soon as they start walking over the pedestrian crossing. The AoA estimation unit, on the other hand, should be capable of providing an AoA estimation of the transmitted signals. Finally, the estimated transmitter direction can be combined with the crossing location in the street to estimate the exact location of the pedestrian.
To validate our proposed system, we conducted an experiment in a street intersection in the smart zone of Antwerp. The intersection consists of 5 pedestrian crossings. We used an IoT transceiver that transmits DASH7 signals. DASH7 standard  is a midrange low power wireless communication standard. It operates over the unlicensed Sub-1 GHz bands. For AoA estimation, we used RTL-array unit . RTL-Array is a cost effective AoA estimation unit that we developed for IoT applications. It operates, as well, over the unlicensed Sub-1 GHz bands.
The results show that the proposed system has the potential to provide an accurate location estimate of the pedestrian that is walking over a pedestrian crossing. The proposed system is also cost effective, low power and does not violate the privacy (comparing to the solutions that are based on deploying cameras). Furthermore, the proposed system can control several pedestrian crossings using a single AoA estimation unit. Therefore, it can be deployed for the smart pedestrian crossing application.
 Alan Bensky (2016). Wireless Positioning Technologies and Applications : second edition. Artech House, London.
 Noori BniLam, Glenn Ergeerts, Dragan Subotic, Jan Steckel , Maarten Weyn. Adaptive Probabilistic Model Using Angle of Arrival Estimation for IoT Indoor Localization, International Conference on Indoor Positioning and Indoor Navigation (IPIN), Sapporo, Japan, 18-21 September 2017.
 Noori BniLam, Dennis Joosens, Jan Steckel, Maarten Weyn. Low Cost AoA Unit for IoT Applications, 13th European Conference on Antennas and Propagation (EuCAP 2019), Krakow, Poland.