LatentWave: JEPA Pretraining for Wireless Foundation Models

• URL: http://arxiv.org/abs/2606.06373v1
• Date: Thu, 04 Jun 2026 16:39:39 GMT
• Title: LatentWave: JEPA Pretraining for Wireless Foundation Models
• Abstract summary: We propose LatentWave, a wireless foundation model pretrained using a Joint-Embedding Predictive Architecture (JEPA)<n>LatentWave learns representations that are more transferable out of the box across diverse downstream tasks.<n>We show that masking geometry introduces a task-dependent inductive bias, while region masking preserves better discriminability for signal classification.
• Score: 7.728643029778198
• License: http://arxiv.org/licenses/nonexclusive-distrib/1.0/
• Abstract: 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.

Related papers

• MultiPathFormer: Towards a Foundation Model for Multipath Wireless Propagation [16.207756414532735]
  We present MultiPathFormer, an autoregressive foundation model that represents each transmitter-receiver link as an ordered sequence of path tokens.<n>We show it can achieve 5.57 m mean localization error, 0.914 top-3 beam accuracy, 0.994 line-of-sight classification accuracy, and 0.561 channel estimation NMSE.
  arXiv  Detail & Related papers  (2026-08-05T17:20:16Z)
• Physics-Informed Domain-Invariant Feature Learning with Autoencoder-Driven Gaussian Clustering for Robust Non-line-of-Sight Scenarios [0.786430477112975]
  Jamming and spoofing pose significant threats to wireless and satellite navigation.<n>We propose a hybrid learning framework that incorporates physics-informed constraints into deep neural networks.<n>We show that the proposed method reduces AoA estimation error by up to 6 in low-exemplar settings.
  arXiv  Detail & Related papers  (2026-06-21T10:42:12Z)
• Learning to Separate RF Signals Under Uncertainty: Detect-Then-Separate vs. Unified Joint Models [53.79667447811139]
  We show that a single deep neural architecture learns to jointly detect and separate when applied directly to the received signal.<n>These findings highlight UJM as a scalable and practical alternative to DTS, while opening new directions for unified separation under broader estimation.
  arXiv  Detail & Related papers  (2026-02-04T15:25:02Z)
• Transmitter Identification via Volterra Series Based Radio Frequency Fingerprint [3.7848713545513317]
  Radio Frequency Fingerprinting (RFF) offers a promising solution as it requires no cryptography and resists spoofing.<n>Existing RFF approaches often lack a unified theory and effective feature extraction.<n>In this work, we model the transmitter as a black box and analyze its impact on transmitted signals.<n> Experiments on a public LoRa dataset show state-of-the-art performance, with over 98% accuracy in static channels and above 90% under multipath and Doppler effects.
  arXiv  Detail & Related papers  (2025-10-22T10:11:41Z)
• Unsupervised Radio Map Construction in Mixed LoS/NLoS Indoor Environments [34.91945910235526]
  This paper aims to recover the data collection trajectory directly from the channel propagation sequence.<n>The proposed method achieves an average localization accuracy of 0.65 meters in an indoor environment.
  arXiv  Detail & Related papers  (2025-10-09T09:53:24Z)
• RF Challenge: The Data-Driven Radio Frequency Signal Separation Challenge [66.33067693672696]
  We address the critical problem of interference rejection in radio-frequency (RF) signals using a data-driven approach that leverages deep-learning methods.<n>A primary contribution of this paper is the introduction of the RF Challenge, which is a publicly available, diverse RF signal dataset.
  arXiv  Detail & Related papers  (2024-09-13T13:53:41Z)
• Generating High Dimensional User-Specific Wireless Channels using Diffusion Models [27.848413167129774]
  This paper introduces a novel method for generating synthetic wireless channel data using diffusion-based models.<n>Our approach employs a conditional denoising diffusion implicit model (cDDIM) framework, effectively capturing the relationship between user location and multi-antenna channel characteristics.<n>We generate synthetic high fidelity channel samples using user positions as conditional inputs, creating larger augmented datasets to overcome measurement scarcity.
  arXiv  Detail & Related papers  (2024-09-05T22:08:28Z)
• A Universal Deep Neural Network for Signal Detection in Wireless Communication Systems [35.07773969966621]
  Deep learning (DL) has been emerging as a promising approach for channel estimation and signal detection in wireless communications. To cope with the dynamic nature of the wireless channel, DL methods must be re-trained on newly non-aged collected data. This paper proposes a novel universal deep neural network (Uni-DNN) that can achieve high detection performance in various wireless environments without retraining the model.
  arXiv  Detail & Related papers  (2024-04-03T11:21:10Z)
• Physical-Layer Semantic-Aware Network for Zero-Shot Wireless Sensing [74.12670841657038]
  Device-free wireless sensing has recently attracted significant interest due to its potential to support a wide range of immersive human-machine interactive applications. Data heterogeneity in wireless signals and data privacy regulation of distributed sensing have been considered as the major challenges that hinder the wide applications of wireless sensing in large area networking systems. We propose a novel zero-shot wireless sensing solution that allows models constructed in one or a limited number of locations to be directly transferred to other locations without any labeled data.
  arXiv  Detail & Related papers  (2023-12-08T13:50:30Z)
• Multi-task Learning Approach for Modulation and Wireless Signal Classification for 5G and Beyond: Edge Deployment via Model Compression [1.218340575383456]
  Future communication networks must address the scarce spectrum to accommodate growth of heterogeneous wireless devices. We exploit the potential of deep neural networks based multi-task learning framework to simultaneously learn modulation and signal classification tasks. We provide a comprehensive heterogeneous wireless signals dataset for public use.
  arXiv  Detail & Related papers  (2022-02-26T14:51:02Z)
• Three-Way Deep Neural Network for Radio Frequency Map Generation and Source Localization [67.93423427193055]
  Monitoring wireless spectrum over spatial, temporal, and frequency domains will become a critical feature in beyond-5G and 6G communication technologies. In this paper, we present a Generative Adversarial Network (GAN) machine learning model to interpolate irregularly distributed measurements across the spatial domain.
  arXiv  Detail & Related papers  (2021-11-23T22:25:10Z)
• Deep Learning Based Antenna Selection for Channel Extrapolation in FDD Massive MIMO [54.54508321463112]
  In massive multiple-input multiple-output (MIMO) systems, the large number of antennas would bring a great challenge for the acquisition of the accurate channel state information. We utilize the neural networks (NNs) to capture the inherent connection between the uplink and downlink channel data sets and extrapolate the downlink channels from a subset of the uplink channel state information. We study the antenna subset selection problem in order to achieve the best channel extrapolation and decrease the data size of NNs.
  arXiv  Detail & Related papers  (2020-09-03T13:38:52Z)
• Harnessing Wireless Channels for Scalable and Privacy-Preserving Federated Learning [56.94644428312295]
  Wireless connectivity is instrumental in enabling federated learning (FL) Channel randomnessperturbs each worker inversions model update while multiple workers updates incur significant interference on bandwidth. In A-FADMM, all workers upload their model updates to the parameter server using a single channel via analog transmissions. This not only saves communication bandwidth, but also hides each worker's exact model update trajectory from any eavesdropper.
  arXiv  Detail & Related papers  (2020-07-03T16:31:15Z)

This list is automatically generated from the titles and abstracts of the papers in this site.

This site does not guarantee the quality of this site (including all information) and is not responsible for any consequences.