LatentWave: JEPA Pretraining for Wireless Foundation Models
This work addresses the need for transferable wireless representations across heterogeneous configurations and tasks, offering a novel pretraining paradigm that avoids low-level bias.
LatentWave introduces a JEPA-based pretraining method for wireless foundation models that predicts masked regions in latent space, outperforming masked reconstruction baselines on four downstream tasks including RF signal classification, 5G NR positioning, beam prediction, and LoS/NLoS classification, with task-dependent masking geometry effects.
Wireless foundation models have emerged as a promising alternative to building separate models for each wireless task. However, existing approaches rely on masked input reconstruction, which can bias representations toward low-level signal details. In this paper, we propose LatentWave, a wireless foundation model pretrained using a Joint-Embedding Predictive Architecture (JEPA) on diverse wireless spectrograms and channel state information (CSI). By predicting masked regions in latent space, LatentWave learns representations that are more transferable out of the box across diverse downstream tasks. The proposed architecture employs per-channel patch embeddings with stochastic channel sampling during pretraining, allowing it to process variable antenna counts and improving usability across heterogeneous wireless configurations. We evaluate LatentWave on four downstream tasks: RF signal classification, 5G NR positioning, beam prediction, and LoS/NLoS classification, comparing against a masked-modeling baseline (WavesFM) pretrained on the same data. Additionally, we show that the masking geometry introduces a task-dependent inductive bias: frequency masking strongly favors channel-related tasks such as positioning and beam prediction, while region masking better preserves discriminability for signal classification.