Ola Janan

Background: The growing complexity of telecommunications networks, fueled by advancements like the Internet of Things (IoT) and 5G, necessitates dynamic and real-time network performance monitoring. Traditional static systems often fail to address challenges related to scalability, adaptability, and response speed in high-demand environments. Integrating artificial intelligence (AI) with unmanned aerial vehicles (UAVs) presents a transformative approach to overcoming these limitations. Objective: This study aims to evaluate the effectiveness of AI-driven drones for real-time network performance monitoring, focusing on key metrics such as latency, signal strength, throughput, and anomaly detection. Methods: A comprehensive framework was developed, employing reinforcement learning (RL) for path planning and a hybrid temporal-spectral anomaly detection (HTS-AD) algorithm. Experimental validation was conducted using 10 UAVs across simulated and real-world environments, collecting over 3.2 million data points. Statistical analyses, including MANOVA and Bayesian regression, were used to evaluate performance. Results: The proposed system demonstrated significant improvements over traditional methods, including a 24.6% increase in anomaly detection accuracy, a 30% reduction in energy consumption, and 99.9% network coverage in high-density UAV deployments. Conclusion: AI-driven drones offer a scalable, efficient, and reliable solution for network monitoring. By addressing limitations of traditional systems, this study establishes a foundation for next-generation telecommunications infrastructure. Future research should focus on real-world deployment and hybrid security models.

  Background: Unmanned Aerial Vehicles (UAVs) utilizing and active interface with 5G networks has become the new frontier to tackling problems of latency and energy efficiency, interference, and resource management. Although prior researches explained the benefits of UAV integrated networks; overall assessment of various parameters and cases is still scarce. Objective: The article seeks to assess the performance of UAV integrated 5G network in terms of latency, power, signal quality, task coordination and coverage optimization and to ascertain the efficiency of optimization algorithms in the improvement of the integrated 5G network. Methods: Emulations were done in MATLAB and NS3 platforms in urban / suburban / emergency call settings. Latency, power consumption, SINR, and completion time were the performance indicator chosen in the paper. Optimization algorithms: Particle Swarm Optimization (PSO), and Genetic Algorithm (GA), and the Multi-Objective Evolutionary Algorithm (MOEA) is evaluated in terms of Convergence time and Solution quality. Results : UAV-aided networks showed 36.7% and 29.2 % improvement in latency and energy consumption, while 33.6 % enhancement in SINR. MOEA offered the best results with 98.3% solution quality, and the PSO being the most convergence oriented. Minor deviations between simulation and real results highlight the need for adaptive mechanisms. Conclusion: The results presented focus on the enough potential of UAV-assisted 5G networks and their potential influence on improving performances in case of different criteria. Further research should focus on successfully implementing and deploying the proposed solutions and broadening the context of study to include 6G technologies.