| Title | : | Scalable Radio Environment Map Estimation for Next Generation Wireless Networks |
| Speaker | : | Subrata Das (IITM) |
| Details | : | Mon, 9 Dec, 2024 4:00 PM @ SSB 334 |
| Abstract: | : | As the demand for high-speed, reliable connectivity continues to grow, future wireless networks are evolving to include innovative technologies for on-demand network access. One such technology is the use of Unmanned Aerial Vehicles (UAVs) as mobile base stations, valuable in scenarios like disaster recovery, rural connectivity, or large-scale public events, where traditional infrastructure may be unavailable. UAV-based networks offer low-cost, rapidly deployable, and flexible solutions for dynamic, high-demand situations. A key challenge in these networks is determining the optimal UAV position for reliable connectivity. This requires estimating the propagation loss for each ground terminal or user equipment (UE) across the aerial space—creating a Radio Environment Map (REM). Existing REM estimation research focuses on improving accuracy through sparse, noisy signal measurements collected by the UAV, often using machine learning techniques such as deep learning. While these methods improve accuracy, the estimated REMs lose relevance as UEs move, requiring the UAV to gather fresh measurements repeatedly. This limits scalability, especially in dynamic scenarios with high UE mobility or large numbers of users. To address these challenges, we propose SkyScale, a novel approach that uses Radio Tomographic Imaging (RTI) to estimate wireless attenuation characteristics of the terrain. Unlike traditional methods, RTI insights are independent of UE locations, enabling SkyScale to generate REMs for any UE placement without needing additional measurements. We evaluate SkyScale with WiFi and LTE testbeds and simulations. Results show a 10× reduction in measurement costs compared to interpolation methods, making SkyScale highly scalable and adaptable to dynamic network conditions, even in environments with high UE mobility. By eliminating repetitive measurements and leveraging terrain-invariant properties, SkyScale offers a transformative solution for UAV-based network deployments, ensuring robust connectivity in ever-changing environments. |
