A Wireless World Model for AI-Native 6G Networks

Ziqi Chen; Yi Ren; Yixuan Huang; Qi Sun; Nan Li; Yuhong Huang; Chih-Lin I; Yifan Li

cs.NIMar 26, 2026

A Wireless World Model for AI-Native 6G Networks

Authors: Ziqi Chen, Yi Ren, Yixuan Huang, Qi Sun, Nan Li, Yuhong Huang, Chih-Lin I, Yifan Li, +1 more

Organizations: China Mobile Research Institute, No. 32 Xuan Wu Men West Street, Beijing, 100053, China.

Abstract

Integrating AI into the physical layer is a cornerstone of 6G networks. However, current data-driven approaches struggle to generalize across dynamic environments because they lack an intrinsic understanding of electromagnetic wave propagation. We introduce the Wireless World Model (WWM), a multi-modal foundation framework predicting the spatiotemporal evolution of wireless channels by internalizing the causal relationship between 3D geometry and signal dynamics. Pre-trained on a massive ray-traced multi-modal dataset, WWM overcomes the data authenticity gap, further validated under real-world measurement data. Using a joint-embedding predictive architecture with a multi-modal mixture-of-experts Transformer, WWM fuses channel state information, 3D point clouds, and user trajectories into a unified representation. Across the five key downstream tasks supported by WWM, it achieves remarkable performance in seen environments, unseen generalization scenarios, and real-world measurements, consistently outperforming SOTA uni-modal foundation models and task-specific models. This paves the way for physics-aware 6G intelligence that adapts to the physical world.

Figures & tables

Figure 1: The workflow of WWM. a, Multi-modal data source for pre-training and evaluation. The ray tracing simulation is performed on Sionna RT, which consisted of channel State Information (CSI), 3D Point Clouds and User Equipment (UE) Trajectory. The Field Measurement is collected outdoor from China Mobile 6G prototype system. b, Pre-training model architecture and pre-training tasks. The WWM is a pre-trained on JEPA, involving an encoder-predictor architecture. Both encoder and predictor are Multi-modal Mixture of Expert Transformer model, trained on 3 self-supervised mask tasks. c, Downstream tasks for validation based on WWM embedding. The representational capabilities of WWM is verified on 4 downstream tasks with simulated data, and its real-world generalization ability is evaluated on CSI frequency-domain prediction based field measurement.

Figure 2: Large-scale multi-modal wireless dataset spanning diverse simulated and real-world environments. Across these environments, we collect multi-modal data including scenario 3D point clouds, user trajectories and time-synchronized CSI for each sample. a, Representative real-world photographs of the selected urban environments used for ray tracing simulation. From top to bottom: Place de l’Étoile (Paris), Forbidden City (Beijing), Munich urban district (Germany), central business district (Beijing), and Wall Street (New York). b, Corresponding 3D scenario models constructed from geographic data and ground signal coverage maps generated in Sionna RT. c, Extracted 3D point clouds of corresponding 3D scenario models. d, Photograph of the real-world outdoor measurement environment used for channel data acquisition, with the base station (BS) location indicated. e, Satellite view of the measurement site, where the yellow cross marks the BS position and the green trapezoid indicates the UE trajectory. f, 3D point clouds reconstructed from the measurement environment. g, BS hardware of the 6G prototype system used for real-world measurements. h, UE device used for outdoor channel data acquisition.

Figure 3: The model and pre-training process. a, WWM employs a Joint Embedding Predictive Architecture (JEPA) to infer masked multi-modal features in latent space. An online encoder processes visible tokens while a predictor estimates masked embeddings, supervised by an Exponential Moving Average (EMA) based momentum encoder to ensure representation stability. b, Multi-modal Mixture of Experts (MMoE). Heterogeneous inputs—CSI, 3D point clouds, and trajectories—are tokenized via domain-specific embedders (Conv3D, Point-net, and MLP). Within each Transformer block, shared self-attention performs global cross-modal reasoning, followed by modality-specific experts (H-FFN, PC-FFN, P-FFN) to preserve physical inductive biases. c, Pre-training masking strategies. Three complementary strategies supervise the model: Fine-grained and coarse CSI masking to extract multi-scale spatio-temporal propagation features. Trajectory masking to capture kinematic dynamics and their interaction with the electromagnetic environment.

Figure 4: Pre-training results. a, The visualizations of CSI reconstruction. 1st graph: Original 16 timestep CSI sample. 2nd graph: Masked CSI sample used as input to the WWM models for fine-grained masking strategy. 3rd graph: Reconstructed CSI sample using masked input. b, The t-SNE Van der Maaten and Hinton (2008) maps show the encoder’s final-layer embeddings across five sampling schemes—randomly sampled token-level embeddings, samples grouped by city, samples grouped by LOS/NLOS conditions, samples grouped by BS and samples grouped by noise levels.

Figure 5: Downstream task and results comparison a, The CSI temporal prediction task details. b, CSI temporal prediction performance comparison with SOTA WFM Wifo and SOTA task-specific model LSTM. c, CSI compression and feedback task details. d, CSI compression and feedback performance comparison with SOTA task-specific models QCR-NET and CR-NET across various compression rate. e, Beam prediction task details. f, Beam prediction across frequency bands performance comparison with SOTA WFM LWM and SOTA task-specific model SPMM. g, User localization task details. h, User localization performance comparison with SOTA task-specific model deep-CNN. i, CSI frequency-domain prediction in real-world measurement task details. j, CSI frequency-domain prediction performance comparison with SOTA WFM Wifo and SOTA task-specific model C-Mixer.

Figure 6: Visualization of CSI frequency-domain prediction based on SRS measurement. Comparison between ground-truth and predicted channel tensors across real, imaginary, magnitude, and phase components for a representative sample. The horizontal axis corresponds to the joint UE antenna and subband dimension, and the vertical axis corresponds to base-station antennas. a, b, Ground-truth real and imaginary parts of the channel tensor. c, d, Corresponding predicted real and imaginary parts produced by the model. e, f, Squared error maps for the real and imaginary components, respectively, showing low reconstruction error across most spatial–frequency elements. g, h, Ground-truth channel magnitudes and phase. i, j, Corresponding predicted channel magnitudes and phase. k, l, Magnitude and Phase error map, indicating small deviations concentrated in limited regions, confirming accurate reconstruction of signal magnitudes and phase.

Symbol	Description	Value
$f_{c}$	Center frequency	2.6 GHz
$\Delta f$	Subcarrier spacing	15 kHz
$F$	Number of subcarriers	96
$N_{\mathrm{sb}}$	Number of subbands	8
$F_{\mathrm{sb}}$	Subcarriers per subband	12
$\Delta t$	Temporal sampling interval	5 ms

Table 1: Extended Data Table 1 — Simulation Dataset Configuration

Symbol	Description	Value
$f_{c}$	Center frequency	6.6 GHz
$\Delta f$	Subcarrier spacing	120 kHz
$N_{\mathrm{sb}}$	Number of subcarriers	3333
$N_{\mathrm{RB}}$	Number of PRBs	264
$N_{\mathrm{RB}sample}$	Number of PRBs per sample	8
$\Delta t$	Temporal sampling interval	10 ms

Table 2: Extended Data Table 2 — Real Measured Dataset Configuration

Scenario ID	City	BS Position	UE Speed	Data Volume
1	Munich	BS0	5 km/h	2048 (Trajectories) $\times$ 16 (Time steps) $\times$ 7 (runs) per scenario
2			30 km/h
3			60 km/h
4		BS1	5 km/h
5			30 km/h
6			60 km/h

Table 3: Extended Data Table 3 — Summary of simulation scenarios and dataset parameters.

Scenario ID	Generalization Type	BS Position	UE Speed	Data Volume
1	Velocity Generalization (Beijing CBD)	BS0	40 km/h	2048 (Trajectories) $\times$ 16 (Time steps) per scenario
2		BS0	70 km/h
3		BS1	40 km/h
4		BS1	70 km/h
5	City Generalization (Wall Street)	BS0	5 km/h
6			30 km/h

Table 4: Extended Data Table 4 — Summary of generalization scenarios and dataset parameters. This portion of the dataset is specifically designed to evaluate the model’s robustness to unseen velocities and urban environments.

Scenario	WWM			WiFo			LSTM
Scenario	T=15	T=16	Avg.	T=15	T=16	Avg.	T=15	T=16	Avg.
CBD	0.796	0.807	0.802	–	–	0.736	0.697	0.674	0.685
Etoile	0.948	0.941	0.944	–	–	0.844	0.896	0.890	0.893
Forbidden City	0.959	0.957	0.958	–	–	0.863	0.861	0.859	0.860
Munich	0.923	0.913	0.918	–	–	0.743	0.879	0.882	0.881
Velocity Generalization	0.767	0.786	0.776	–	–	0.677	0.694	0.671	0.682

Table 5: Extended Data Table 5 — CSI temporal prediction performance of WWM compared to other baselines across in-distribution and generalization scenarios. A Metric of SGCS of predicted CSI is listed

Model	Scenario	1/1024	1/512	1/256	1/128
WWM	CBD	0.6516	0.6666	0.6610	0.7522
	Etoile	0.8027	0.8027	0.8460	0.9556
	Forbidden City	0.7901	0.8014	0.8244	0.9240
	Munich	0.7704	0.7652	0.7995	0.9467
	Velocity Generalization	0.6468	0.6623	0.6598	0.7454
	City Generalization	0.6214	0.6187	0.6535	0.8961

Table 6: Extended Data Table 6 — Comprehensive comparison of CSI compression and feedback performance (SGCS) across varying compression ratios and scenarios. Performance is evaluated for WWM and baselines (QCR-NET, CR-NET) under in-distribution and generalization regimes.

Scenario	WWM		LWM		SPMM
Scenario	Top-1 Acc	SE ratio	Top-1 Acc	SE ratio	Top-1 Acc	SE ratio
CBD	0.983	1.000	0.927	0.995	0.902	0.987
Etoile	0.917	0.980	0.836	0.944	0.778	0.931
Forbidden City	0.906	0.981	0.854	0.970	0.828	0.963
Munich	0.915	0.984	0.836	0.956	0.767	0.926
Velocity Generalization	0.978	0.999	0.926	0.995	0.857	0.972

Table 7: Extended Data Table 7 — beam prediction performance of WWM compared to other baselines across in-distribution and generalization scenarios. The Metrics of Top-1 accuracy of predicted DFT codeword and their respective beam gain ratio are used

Scenario	WWM	deep-CNN
CBD	1.212743	2.2889
Etoile	2.982986	5.2257
Forbidden City	2.868211	6.0771
Munich	2.718476	4.8768
Velocity Generalization	1.226949	1.9978

Table 8: Extended Data Table 8 — User localization performance of WWM compared to other baselines across in-distribution and generalization scenarios. A Metric of mean average error of 2D distance is used

Scenario	WWM			WWM-Unimodal
Scenario	T=15	T=16	Avg.	T=15	T=16	Avg.
CBD	0.796	0.807	0.802	0.687	0.687	0.687
Etoile	0.948	0.941	0.944	0.935	0.922	0.928
Forbidden City	0.959	0.957	0.958	0.923	0.912	0.918
Munich	0.923	0.913	0.918	0.901	0.879	0.890
Velocity Generalization	0.767	0.786	0.776	0.653	0.662	0.657

Table 9: Extended Data Table 9 — Ablation study comparing the full multimodal WWM with its unimodal variant across scenarios, reporting per-timestep SGCS results (

T{=}15

T{=}16

) and their average.

Scenario	WWM		WiFo		C-Mixer
Scenario	NMSE	SGCS	NMSE	SGCS	NMSE	SGCS
PRB4	0.212	0.932	0.252	0.922	0.296	0.918
PRB5	0.218	0.918	0.261	0.895	0.308	0.892
PRB6	0.260	0.914	0.309	0.882	0.349	0.875
PRB7	0.243	0.902	0.287	0.858	0.334	0.843
Avg.	0.233	0.917	0.277	0.889	0.322	0.882

Table 10: Extended Data Table 10 — CSI frequency-domain prediction performance of WWM compared to other baselines by metric of NMSE and SGCS.

Explore similar work

Physical-AI: From Channel Awareness to Environmental Intelligence in 6G Wireless Networks
May 12, 2026Farooque Hassan Kumbhar, Kapal Dev, Sunder Ali Khowaja +36G Networks Wireless Communications
Augmented Cognition Meta-communications ERC Research Center, Korea University, Korea · Department of Computer Science, Munster Technological University (MTU), Ireland · School of Computing, Faculty of Engineering and Computing, Dublin City University, Ireland +4
AirFM-DDA: Air-Interface Foundation Model in the Delay-Doppler-Angle Domain for AI-Native 6G
Apr 19, 2026Kejia Bian, Meixia Tao, Jianhua Mo +2Wireless Foundation Models Representation Learning
School of Information Science and Electronic Engineering, Shanghai Jiao Tong University, Shanghai, 200240, China · Shanghai Innovation Institute, Shanghai 200231, China
Towards CSI-Native Foundation Models: A Channel-Adaptive Roadmap for 6G
Jun 12, 2026Chenyu Zhang, Xinchen Lyu, Chenshan Ren +2Wireless Foundation Models Representation Learning
National Engineering Research Center for Mobile Network Technologies, Beijing University of Posts and Telecommunications, Beijing 100876, China · Key Laboratory of Ethnic Language Intelligent Analysis and Security Governance of MOE, Minzu University of China, Beijing 100081, China · China Telecom Corporation Limited Gansu Branch, Gansu 730000, China

May 12, 2026J/K move · Enter open · S save

Physical-AI: From Channel Awareness to Environmental Intelligence in 6G Wireless Networks

Farooque Hassan Kumbhar, Kapal Dev, Sunder Ali Khowaja, Alexandros-Apostolos A. Boulogeorgos+2

Augmented Cognition Meta-communications ERC Research Center, Korea University, Korea · Department of Computer Science, Munster Technological University (MTU), Ireland · School of Computing, Faculty of Engineering and Computing, Dublin City University, Ireland+4

6G Networks Wireless Communications

Open paper AskPDF arXiv

Conventional wireless networks rely on instantaneous channel state information (CSI) and react to channel variations without explicitly modeling the physical environment, limiting their ability to handle blockage, mobility, and interference in dynamic deployments. Paradigms such as Integrated Sensing and Communication (ISAC) add sensing capabilities but lack explicit environment modeling and decision-making. In this article, we propose Physical-AI: a new architecture for environment-aware wireless networking, where radio signals enable sensing, modeling, and interaction with the environment in addition to data transmission. The framework proposes a self-supervised spatiotemporal radio foundation model for transforming distributed radio observations into a shared latent environmental representation. Multiple inference heads operate on this representation to estimate key environmental properties, including blockage, user distribution, mobility dynamics, and interference structure. A task-specific neural decision layer maps this representation to proactive, context-aware control actions. By integrating perception, world modeling, and decision-making in a closed loop, the proposed framework goes beyond ISAC and establishes Physical-AI as a promising architecture for intelligent 6G systems. Simulation results show that the proposed predictive framework reduces outage probability and blockage-response latency, particularly under increasing beam-switching delays.

Abstract

Explore similar work

Physical-AI: From Channel Awareness to Environmental Intelligence in 6G Wireless Networks

AirFM-DDA: Air-Interface Foundation Model in the Delay-Doppler-Angle Domain for AI-Native 6G

Towards CSI-Native Foundation Models: A Channel-Adaptive Roadmap for 6G