Browsing by Author "Jalali, Erfan"

Filter results by typing the first few letters
Now showing 1 - 1 of 1
  • Thumbnail Image
    ItemOpen Access
    Dynamic user motion prediction using advanced Kalman filtering in 5G mmWave systems
    (Faculty of Graduate Studies and Research, University of Regina, 2025-04) Jalali, Erfan; Paranjape, Raman; Wang, Zhanle (Gerald)
    Achieving reliable communication in high-data-rate applications and densely populated environments remains a significant challenge for next-generation wireless networks. Millimeterwave (mmWave) technology offers substantial bandwidth and data rate advantages but necessitates precise beam steering to maintain connectivity. The dynamic nature of mobile environments, coupled with diverse user behaviors and trajectories, complicates this task. Traditional beamforming approaches struggle to adapt to such scenarios, often leading to signal degradation, connectivity drops, inefficient resource utilization and more power consumption. This research presents an advanced framework that integrates historical user trajectories through beam sweeping, user tracking via sensor-based information combined with beam sweeping, and Kalman Filter-based prediction to address these challenges effectively. By leveraging key measurements such as Angle of Arrival (AoA), Received Signal Strength (RSS) and Signal-to-Noise-and-Interference Ratio (SINR), the system dynamically tracks and predicts user locations based on data gathered by beam sweeping. This research introduces a sophisticated framework that leverages historical user movement patterns, collected via IMEI-based tracking with beam sweeping over time, to define prototype trajectories. These trajectories are subsequently refined using advanced trajectory identification algorithms. Additionally, multiple sensors installed along the pathway provide approximate user positions as they move along their trajectories. The Kalman Filter, particularly its nonlinear models (Advanced Kalman Filters), significantly enhances prediction accuracy, enabling real-time tracking and prediction of user movements. This capability facilitates future beam steering and adaptive beamforming to optimize signal transmission. The proposed framework addresses several critical challenges. It effectively manages dynamic user behaviors and nonlinear trajectories through Kalman Filter-based prediction. By dynamically allocating channels to users, even in overlapping paths, it ensures efficient resource management, reducing interference and enhancing connectivity. Moreover, the integration of trajectorybased tracking preemptively predicts user positions, allowing for seamless beam adjustments, particularly at curves and cell edges, thereby preventing connectivity drops and maintaining robust communication links. The simulation results demonstrate the robustness of the proposed approach in complex indoor environments, showcasing significant improvements in user tracking, enhanced system awareness of user motion, accurate estimations within an acceptable range, and overall system reliability. The results of this study serve as a valuable asset for the adaptive beamforming framework, not only enhancing RSS and SINR but also reducing power consumption by activating idle-mode antennas. These improvements collectively enhance user experience and ensure seamless connectivity in high-density environments. This study offers a scalable and efficient solution for next-generation wireless networks in diverse indoor scenarios such as smart cities, shopping malls, large public venues and mining tunnels. The findings provide a strong foundation for the development of robust and reliable communication systems in the era of mmWave technology.