# Towards a Wireless Physical-Layer Foundation Model: Challenges and Strategies

arXiv: https://arxiv.org/abs/2403.12065 · HTML: https://mlsanity.com/paper/2403.12065/

## Key facts

- Cited by 11 papers in the map (data through 14 Oct 2026).
- First posted on arXiv 9 Feb 2024.

## Built on by (most cited first, 11 of 11)

- [WirelessLLM: Empowering Large Language Models Towards Wireless Intelligence](https://mlsanity.com/paper/2405.17053/) (2024)
- [A Multi-Task Foundation Model for Wireless Channel Representation Using Contrastive and Masked Autoencoder Learning](https://mlsanity.com/paper/2505.09160/) (2025)
- [6G WavesFM: A Foundation Model for Sensing, Communication, and Localization](https://mlsanity.com/paper/2504.14100/) (2025)
- [Building 6G Radio Foundation Models with Transformer Architectures](https://mlsanity.com/paper/2411.09996/) (2024)
- [RF Challenge: The Data-Driven Radio Frequency Signal Separation Challenge](https://mlsanity.com/paper/2409.08839/) (2024)
- [EMind: A Foundation Model for Multi-task Electromagnetic Signals Understanding](https://mlsanity.com/paper/2508.18785/) (2025)
- [Multi-Modal Sensor Fusion for Proactive Blockage Prediction in mmWave Vehicular Networks](https://mlsanity.com/paper/2507.15769/) (2025)
- [Large Language Models Can Perform Automatic Modulation Classification via Discretized Self-supervised Candidate Retrieval](https://mlsanity.com/paper/2510.00316/) (2025)
- [MultiPathFormer: Towards a Foundation Model for Multipath Wireless Propagation](https://mlsanity.com/paper/2608.05076/) (2026)
- [A Comprehensive Survey of Wireless Foundation Models for AI-Native 6G Networks](https://mlsanity.com/paper/2608.14694/) (2026)
- [Wireless Foundation Models: State-of-the-Art and Open Challenges](https://mlsanity.com/paper/2609.04707/) (2026)

## Abstract (arXiv)

Artificial intelligence (AI) plays an important role in the dynamic landscape of wireless communications, solving challenges unattainable by traditional approaches. This paper discusses the evolution of wireless AI, emphasizing the transition from isolated task-specific models to more generalizable and adaptable AI models inspired by recent successes in large language models (LLMs) and computer vision. To overcome task-specific AI strategies in wireless networks, we propose a unified wireless physical-layer foundation model (WPFM). Challenges include the design of effective pre-training tasks, support for embedding heterogeneous time series and human-understandable interaction. The paper presents a strategic framework, focusing on embedding wireless time series, self-supervised pre-training, and semantic representation learning. The proposed WPFM aims to understand and describe diverse wireless signals, allowing human interactivity with wireless networks. The paper concludes by outlining next research steps for WPFMs, including the integration with LLMs.