Planning and Deployment of Wireless Networks: A Data-Driven Machine Learning and Optimization Framework Based on Urban Mesh and 5G Networks

Wireless networks are essential infrastructure for applications from industrial automation to smart cities. Despite varying architectures—mobile, sensor, and mesh networks—they face common challenges in urban deployment, such as signal propagation, connectivity, and energy efficiency. This thesis proposes data-driven approaches using machine learning and optimization to address key gaps in the literature, particularly in accurate signal strength prediction and relay node placement. Three interrelated studies are presented. The first applies a machine learning model to predict the Received Signal Strength Indicator (RSSI) in urban mesh networks, significantly outperforming traditional methods. The second provides a review of feature selection for signal prediction across network types, enabling more accurate and efficient models. The third proposes an optimization model for relay node placement, balancing connectivity, energy efficiency, cost, and fault tolerance. Overall, these contributions advance the state-of-the-art on planning and deployment of wireless networks in real-world urban environments.