This challenge leverages four provided CSV datasets containing network performance metrics for 2,880 beams across 30 base stations. Each base station consists of 3 cells with 32 beams, with data recorded hourly. These datasets encompass a five-week period with data recorded at hourly intervals (as illustrated in Figure 2). These datasets are traffic_DLThpVol.csv, traffic_DLThpTime.csv, traffic_MR_number.csv, and traffic_DLPRB.csv. We remind the participants that the objective is to forecast future values of traffic volume (DLThpVol).

Each dataset corresponds to a specific network performance metric:

• traffic_DLThpVol.csv: represents throughput volume.
• traffic_DLThpTime.csv: represents throughput time.
• traffic_ DLPRB.csv: represents Physical Resource Block (PRB) utilization.
• traffic_MR_number.csv: represents user count.

Besides the dataset, complementary material will be provided to participants, including references, detailed instructions, and guidelines.

[1] Hyndman, R.J., Athanasopoulos, G. “Forecasting: Principles and Practice.” OTexts, 2021.

[2] Liu, Yong, et al. “iTransformer: Inverted Transformers Are Effective for Time Series Forecasting.” ICLR, 2024.

[3] Nie, Yuqi, et al. “A Time Series is Worth 64 Words: Long-term Forecasting with Transformers.” ICLR, 2023.

[4] Zeng, Ailing, et al. “Are transformers effective for time series forecasting?” AAAI, 2023.

[5] Vaswani, Ashish, et al. “Attention is all you need.” NeurIPS, 2017.