CoBA: Integrated Deep Learning Model for Reliable Low-Altitude UAV Classification in mmWave Radio Networks

Uncrewed Aerial Vehicles (UAVs) are increasingly used in civilian and industrial applications, making secure low-altitude operations crucial. In dense mmWave environments, accurately classifying low-altitude UAVs as either inside authorized or restricted airspaces remains challenging, requiring models that handle complex propagation and signal variability. This paper proposes a deep learning model, referred to as CoBA, which stands for integrated Convolutional Neural Network (CNN), Bidirectional Long Short-Term Memory (BiLSTM), and Attention which leverages Fifth Generation (5G) millimeter-wave (mmWave) radio measurements to classify UAV operations in authorized and restricted airspaces at low altitude. The proposed CoBA model integrates convolutional, bidirectional recurrent, and attention layers to capture both spatial and temporal patterns in UAV radio measurements. To validate the model, a dedicated dataset is collected using the 5G mmWave network at TalTech, with controlled low altitude UAV flights in authorized and restricted scenarios. The model is evaluated against conventional ML models and a fingerprinting-based benchmark. Experimental results show that CoBA achieves superior accuracy, significantly outperforming all baseline models and demonstrating its potential for reliable and regulated UAV airspace monitoring.

Reconfigurable Intelligent Surfaces in Dynamic Rich Scattering Environments: BiLSTM-Based Optimization for Accurate User Localization

The integration of reconfigurable intelligent surfaces (RIS) in wireless environments offers channel programmability and dynamic control over propagation channels, which is expected to play a crucial role in sixth generation (6G) networks. The majority of RIS-related research has focused on simpler, quasi-free-space conditions, where wireless channels are typically modeled analytically. However, many practical localization scenarios unfold in environments characterized by rich scattering that also change over time. These dynamic and complex conditions pose significant challenges in determining the optimal RIS configuration to maximize localization accuracy. In this paper, we present our approach to overcoming this challenge. This paper introduces a novel approach that leverages a bidirectional long-short term memory (biLSTM) network, trained with a simulator that accurately reflects wave physics, to capture the relationship between wireless channels and the RIS configuration under dynamic, rich-scattering conditions. We use this approach to optimize RIS configurations for enhanced user equipment (UE) localization, measured by mean squared error (MSE). Through extensive simulations, we demonstrate that our approach adapts RIS configurations to significantly improve localization accuracy in such dynamically changing rich scattering environments.

Role of Reconfigurable Intelligent Surfaces in 6G Radio Localization: Recent Developments, Opportunities, Challenges, and Applications

Reconfigurable intelligent surfaces (RISs) are seen as a key enabler low-cost and energy-efficient technology for 6G radio communication and localization. In this paper, we aim to provide a comprehensive overview of the current research progress on the RIS technology in radio localization for 6G. Particularly, we discuss the RIS-assisted radio localization taxonomy and review the studies of RIS-assisted radio localization for different network scenarios, bands of transmission, deployment environments, as well as near-field operations. Based on this review, we highlight the future research directions, associated technical challenges, real-world applications, and limitations of RIS-assisted radio localization.