A MIMO Wireless Channel Foundation Model via CIR-CSI Consistency

TL;DR

This work addresses the need for robust, generalizable representations of wireless channels across diverse scenarios. It introduces CSI-CLIP, a CLIP-like, dual-pathway framework that treats Channel State Information ($CSI$) and Channel Impulse Response ($CIR$) as aligned multimodal data and learns shared embeddings through a single contrastive objective with a temperature parameter. The approach yields substantial downstream benefits, including an average $22\%$ improvement in positioning and about $1\%$ in beam management over supervised baselines, and demonstrates cross-scenario generalization on DeepMIMO and Sionna datasets with strong robustness to frequency variations. By enabling effective ISAC integration through CIR-$CSI$ consistency, CSI-CLIP opens new research directions in multi-modal wireless perception and robust MIMO processing, while highlighting areas for future work on antenna and subcarrier generalization and linear probing.

Abstract

In the field of artificial intelligence, self-supervised learning has demonstrated superior generalization capabilities by leveraging large-scale unlabeled datasets for pretraining, which is especially critical for wireless communication models to adapt to a variety of scenarios. This paper innovatively treats Channel State Information (CSI) and Channel Impulse Response (CIR) as naturally aligned multi-modal data and proposes the first MIMO wireless channel foundation model, named CSI-CLIP. By effectively capturing the joint representations of both CIR and CSI, CSI-CLIP exhibits remarkable adaptability across scenarios and robust feature extraction capabilities. Experimental results show that in positioning task, CSI-CLIP reduces the mean error distance by 22%; in beam management task, it increases accuracy by 1% compared to traditional supervised methods, as well as in the channel identification task. These improvements not only highlight the potential and value of CSI-CLIP in integrating sensing and communication but also demonstrate its significant advantages over existing techniques. Moreover, viewing CSI and CIR as multi-modal pairs and contrastive learning for wireless channel foundation model open up new research directions in the domain of MIMO wireless communications.

Figures (2)

• Figure 1: Visualize for the CSI and CIR in MIMO topology.
• Figure 2: Architecture of the proposed CSI-CLIP.