NextG Base Station Placement and Failure Mitigation

The deployment of resilient NextG networks necessitates a dual-focus approach: optimal base stations (BS) placement and efficient operational maintenance. This paper presents an integrated two-stage framework that systematically addresses both objectives. Stage one introduces a GPU-accelerated methodology for BS placement, employing K-Means clustering for initial seeding, followed by a novel hybrid metaheuristic - Fuzzy Role-based Foraging Optimization (FRFO) - to maximize population coverage. Performance evaluation is conducted under the 3GPP Rural Macrocell (RMa) propagation model. FRFO is rigorously benchmarked against six established algorithms (including K-means). Results indicate that FRFO achieves significantly faster convergence and enhanced coverage performance compared with state-of-the-art algorithms. Stage two builds on the optimal BS deployment by introducing a novel, density-aware pathfinding algorithm for dispatching autonomous ground robots to repair failed BSs. This algorithm accounts for both distance and turning delays and is evaluated against standard Dijkstra and A∗ methods. Simulation results validate that the proposed framework offers a comprehensive solution across the network lifecycle - from initial deployment to maintenance - achieving enhanced coverage and reduced downtime, thereby supporting the robust operation of next-generation wireless networks. © 2026 IEEE.