A Wireless Foundation Model for Multi-Task Prediction

--With the growing complexity and dynamics of the mobile communication networks, accurately predicting key system parameters, such as channel state information (CSI), user location, and network traffic, has become essential for a wide range of physical (PHY)-layer and medium access control (MAC)-layer tasks. Although traditional deep learning (DL)- based methods have been widely applied to such prediction tasks, they often struggle to generalize across different scenarios and tasks. In response, we propose a unified foundation model for multi-task prediction in wireless networks that supports diverse prediction intervals. The proposed model enforces univariate decomposition to unify heterogeneous tasks, encodes granularity for interval awareness, and uses a causal Transformer backbone for accurate predictions. Additionally, we introduce a patch masking strategy during training to support arbitrary input lengths. After trained on large-scale datasets, the proposed foundation model demonstrates strong generalization to unseen scenarios and achieves zero-shot performance on new tasks that surpass traditional full-shot baselines. HE advent of 6G communications [1] has made wireless systems more intricate, featuring ultra-dense deployments, diverse service demands, and highly dynamic environments. Efficient execution of physical (PHY) and medium access control (MAC)-layer tasks require accurate and timely knowledge of the surrounding communication environment. Key parameters of interest include channel state information (CSI) [2], user locations [3], mobile traffic at the base station (BS) [4], etc. However, these parameters fluctuate rapidly over time, making real-time estimation and feedback particularly challenging. As a result, accurately predicting these variables has become essential for enabling a wide range of downstream communication tasks.