Projects.

With the recent advancements in wireless technologies, forecasting electromagnetic field (EMF) exposure has become increasingly critical to enable proactive network spectrum and power allocation, and network deployment planning. In this paper, we propose EMForecaster, a novel deep learning time series forecasting architecture which employs patching to process temporal patterns at multiple scales, complemented by reversible instance normalization and mixing operations along both temporal and patch dimensions for efficient feature extraction. We augment EMForecaster with a conformal prediction, a distribution-free uncertainty quantification mechanism, to enhance confidence of model forecasts. In particular, conformal prediction ensures that the ground truth lies within a prediction interval with target error rate $\alpha$, where $1-\alpha$ is referred to as coverage. However, a trade-off exists, as increasing coverage often results in wider prediction intervals. To address this challenge, we propose a new metric referred to as Trade-off Score, that balances the trustworthiness of the forecast (i.e., coverage) and the width of prediction interval. Our empirical evaluation demonstrates that EMForecaster achieves superior performance across diverse EMF datasets and forecast horizons. In point forecasting, EMForecaster outperforms current state-of-the-art DL approaches, 53.97% better than the Transformer architecture and 38.44% over the average baseline. For conformal time series forecasting, EMForecaster provides an excellent balance between prediction interval width and coverage, reaching comparable performance to DLinear, surpassing the average baseline by 24.73% and the Transformer by 49.17%.