AIU4 – Localization
Friday, 6 June 2025, 9:00-10:30, room 1.F
Session Chair: Krzysztof Cichoń (Poznan Univ. Technology, PL)
Joint Mobile IAB Node Positioning and Scheduler Selection in Locations with Significant Obstacles
Paulo Furtado Correia (Universidade do Porto & INESCTEC, Portugal); André Coelho (INESC TEC, Portugal); Manuel Ricardo (Universidade do Porto & INESC TEC, Portugal)
Integrated Access and Backhaul (IAB) in cellular networks combines access and backhaul within a wireless infrastructure reducing reliance on fibre-based backhaul. This enables flexible and more cost-effective network expansion, especially in hard-to-reach areas. Positioning a mobile IAB node (MIAB) in a seaport environment, in order to ensure on-demand, resilient wireless connectivity, presents unique challenges due to the high density of User Equipments (UEs) and potential shadowing effects caused by obstacles. This paper addresses the problem of positioning MIABs within areas containing UEs, fixed IAB donors (FIABs), and obstacles. Our approach considers user associations and different types of scheduling, ensuring MIABs and FIABs meet the capacity requirements of a special team of served UEs, while not exceeding backhaul capacity. With a Genetic Algorithm solver, we achieve capacity improvement gains, by up to 200% for the 90th percentile, particularly during emergency capacity demands.
Using Machine Learning to Localize BLE Devices on a Single Anchor
Samuel Leitch and Qasim Zeeshan Ahmed (University of Huddersfield, United Kingdom (Great Britain)); Jaron Fontaine (Ghent University – IMEC, Belgium); Ben Van Herbruggen (Ghent University & Imec, Belgium); Adnan Shahid (Gent University – Imec, Belgium); Eli De Poorter (Ghent University & Imec, Belgium); Pavlos Lazaridis (University of Huddersfield, United Kingdom (Great Britain))
Indoor localization using Bluetooth Low Energy (BLE) technology can be accomplished by a variety of methods. One which provides some appreciable benefits is the single-anchor solution, which allows for low-cost deployments of systems. In this paper, five different methods of single-anchor localization have been investigated including different methods of angle of arrival determination and distance estimation. It was found that the best-performing method of single-anchor localization was a dedicated Machine Learning algorithm whose output is the location of the target device. Once a Kalman filter was applied to it, it achieved a mean distance error of 0.34m on the test scenario.
Zero-Energy Devices for 6G: First Real-Time Indoor Localization Thanks to Ambient Backscattering of Commercial 4G UEs
Ahmed ElSanhoury, Sr (Orange Innovation Centre, Egypt); Islam Galal, Khaled Alkady, Aml Elkhodary and Hashem Elbiali (Orange Innovation Center, Egypt); Ayman Mostafa Hassan (Benha University, Egypt); Dinh-Thuy Phan-Huy (Orange, France)
Traditional approaches to indoor localization of smartphones, such as Wi-Fi signal strength, Bluetooth low energy (BLE) localization techniques, face limitations like energy consumption and costly infrastructure deployment. This paper presents a novel indoor localization system, based zero-energy devices (ZED) which are energy-autonomous because powered by light, and based on ambient backscatter communication (AmBC) technique which re-uses ambient mobile networks infrastructures and waves. Uplink pilot signals broadcasted by user equipements (UEs) are backscattered ZEDs close by, and base stations (BS) detect the ZED message in the received UL pilot signal. We implement two flavors of our localization system. In the first flavor, ZEDs are fixed with known locations in rooms and the location of the UE (a smartphone) carried by a moving user is tracked. In the second flavor, the UEs (4G modems) are fixed and in known locations, and the location of the flexible ZED worn by a moving user is tracked. We demonstrate the two flavors, in real-time, in a real indoor environment, with 4th Generation (4G) commercial UEs (4G smartphone and 4G modem) and a 4G software defined radio base station. We achieve room-based detection accuracy.
Deep Learning-Based Data Fusion of 6G Sensing and Inertial Information for Target Positioning: Experimental Validation
Karthik Muthineni (Robert Bosch GmbH, Germany & Universitat Politècnica de Catalunya, Spain); Alexander Artemenko (Robert Bosch GmbH, Germany); Artjom Grudnitsky (Nokia Bell Labs, Germany); Josep Vidal and Montse Nájar (Universitat Politècnica de Catalunya, Spain)
The sixth-generation (6G) cellular technology will be deployed with a key feature of Integrated Sensing and Communication (ISAC), allowing the cellular network to map the environment through radar sensing on top of providing communication services. In this regard, the entire network can be considered as a sensor with a broader Field of View (FoV) of the environment, assisting in both the positioning of active and detection of passive targets. On the other hand, the non-3GPP sensors available on the target can provide additional information specific to the target that can be beneficially combined with ISAC sensing information to enhance the overall achievable positioning accuracy. In this paper, we first study the performance of the ISAC system in terms of its achievable accuracy in positioning the mobile target in an indoor scenario. Second, we study the performance gain achieved in the ISAC positioning accuracy after fusing the information from the target’s non-3GPP sensors. To this end, we propose a novel data fusion solution based on the deep learning framework to fuse the information from ISAC and non-3GPP sensors. We validate our proposed data fusion and positioning solution with a real-world ISAC Proof-of-Concept (PoC) as the wireless infrastructure, an Automated Guided Vehicle (AGV) as the target, and the Inertial Measurement Unit (IMU) sensor on the target as the non-3GPP sensor. The experimental results show that our proposed solution achieves an average positioning error of 3 cm, outperforming the considered baselines.
Optimizing Access Point Placement Using GNN for Enhancing Indoors UE Localization
Gerasimos Papanikolaou-Ntais (NCSR Demokritos, Greece); Vasiliki Rentoula (National Centre for Scientific Research Demokritos, Greece); Ilias Alexandropoulos (National Centre of Scientific Research Demokritos, Greece); George Makropoulos (NCSR Demokritos & National and Kapodistrian University of Athens, Greece); Harilaos Koumaras (NCSR Demokritos, Greece)
Indoor WiFi localization has become a critical aspect of modern wireless networks, enabling applications such as indoor navigation and location-based services. While most machine learning models for localization rely on static access point (AP) deployment, this paper introduces a novel approach that incorporates dynamic deployment. It uses Graph Neural Networks (GNNs) to optimize AP placement, thereby enhancing the accuracy of indoor User Equipment (UE) localization. This approach is built on top of NS-3 simulations, which utilize realistic indoor topologies and combine WiFi and mmWave APs. The GNN is trained to learn effective representations of AP positions. Simulations in NS-3 generate realistic network traffic datasets, showing that GNN-optimized AP placements significantly improve localization accuracy compared to baseline models.
