CELL_DATASET_FOR_RL

This dataset provides anonymized, time-series Key Performance Indicator (KPI) logs from a four-site LTE network cluster (Mathurapur region) that underwent a real-time parameter optimization trial. The data records network performance metrics prior to, and during, the deployment of a Q-learning-based Reinforcement Learning (RL) agent designed to autonomously manage cell transmit power and antenna electrical tilt. The primary value of this dataset is its empirical nature, reflecting the noisy, multi-objective, and complex correlation landscape of a real-world cellular environment. The data is essential for validating and reproducing the RL agent's performance in achieving simultaneous improvements in competing KPIs such as throughput, interference, and call stability