Design and Implementation of UE Detection and Localization Algorithms Using Control Signals in Drone-Assisted 5G NR for Emergency Response

In modern disaster response operations, the ability to locate users by their mobile phones without the need to be connected to terrestrial infrastructure has become a critical capability. This thesis investigates the use of Unmanned Aerial Vehicle (UAV) equipped with a 5G New Radio (NR) base station to detect and localize User Equipment (UE) based only on control signals. Using an OpenAirInterface-based testbed (OAIBox), we analyze the detection and localization potential of 5G NR uplink reference signals, such as the Physical Random Access Channel (PRACH) and the Sounding Reference Signal (SRS). The proposed approach exploits the fact that even idle or disconnected UEs periodically transmit PRACH preambles during the initial access procedure, which can be passively received by the UAV. Different algorithms using Received Signal Strength (RSS) are developed to improve the reliability in Line-of-sight propagation (LoS) and Non-line-of-sight propagation (NLoS) conditions. The practical implementation involves configuring an OAIBox to capture and log physical-layer data in real time during UAV flights. The system design accounts for drone flight constraints such as power consumption, hovering time, and real-time signal processing limits. The pro- posed localization framework is validated through simulations, demonstrating the feasibility of infrastructure-independent UE detection in emergency or remote environments. An important contribution of this thesis is the design of an adaptive UAV flight strategy and signal-based UE localization algorithms. The flight strategy optimizes the drone trajectory for improved spatial coverage and measurement quality, while the proposed algorithms are tailored to operate reliably in both LoS and NLoS conditions. Simulations demonstrated that the proposed system can reliably detect user devices and achieve meter-level positioning accuracy under LoS conditions, while maintaining room-level accuracy in NLoS scenarios. This work lays the foundation for a fully autonomous drone-based localization system that can assist rescue teams in post-disaster scenarios, where conventional network access is unavailable.