Speaker
Description
In this paper, we propose a big data generation method that considers real-world 5G base station operational scenarios. The proposed method generates a big data that mimics beamforming, base station deployment conditions, and various operational scenarios. Since the generated dataset reflects real-world conditions, it can be used to train recently developed AI-based prediction techniques for 5G base station evaluation. The proposed method consists of two main processes: the generation of radiation patterns and the execution of ray-tracing simulations using the generated radiation patterns. To facilitate big data generation, each process is automated using MATLAB and Python. We will demonstrate the diversity of the generated data using the proposed method in the results section, verifying that it enables the generation of datasets that reflect various real-world 5G operational scenarios.
