Large AI Model for Delay-Doppler Domain Channel Prediction in 6G OTFS-Based Vehicular Networks

TL;DR

The paper tackles the difficulty of predicting highly dynamic vehicular channels by adopting a delay-Doppler domain (DD) representation via OTFS, which yields a sparse and stable channel description that aligns with physical propagation. It then casts channel prediction as forecasting time-series parameters (|g(i,t0)|, phase differences, delays, and Doppler) of DD-domain sub-paths, using a large transformer-based AI model named Timer with zero-shot capabilities and optional fine-tuning on vehicular data. Experiments on V2I highway scenarios show that fine-tuned Timer consistently outperforms LSTM/GRU baselines and zero-shot large AI models in predicting DD-domain parameters and total path loss, while zero-shot deployment remains advantageous for low-cost predictions. The results indicate strong potential for robust, low-latency vehicular communications, and the approach generalizes to other high-mobility settings such as UAVs and satellite-terrestrial links; future work includes link adaptation aided by channel predictions.

Abstract

Channel prediction is crucial for high-mobility vehicular networks, as it enables the anticipation of future channel conditions and the proactive adjustment of communication strategies. However, achieving accurate vehicular channel prediction is challenging due to significant Doppler effects and rapid channel variations resulting from high-speed vehicle movement and complex propagation environments. In this paper, we propose a novel delay-Doppler (DD) domain channel prediction framework tailored for high-mobility vehicular networks. By transforming the channel representation into the DD domain, we obtain an intuitive, sparse, and stable depiction that closely aligns with the underlying physical propagation processes, effectively reducing the complex vehicular channel to a set of time-series parameters with enhanced predictability. Furthermore, we leverage the large artificial intelligence (AI) model to predict these DD-domain time-series parameters, capitalizing on their advanced ability to model temporal correlations. The zero-shot capability of the pre-trained large AI model facilitates accurate channel predictions without requiring task-specific training, while subsequent fine-tuning on specific vehicular channel data further improves prediction accuracy. Extensive simulation results demonstrate the effectiveness of our DD-domain channel prediction framework and the superior accuracy of the large AI model in predicting time-series channel parameters, thereby highlighting the potential of our approach for robust vehicular communication systems.

Figures (8)

• Figure 1: Quasi deterministic channel modeling for high-mobility vehicular networks.
• Figure 2: Delay-Doppler domain channel prediction with the large AI model.
• Figure 3: Large AI model with Transformer as backbone.
• Figure 4: Loss curves of fine-tuning for the large AI model.
• Figure 5: CDF of prediction errors for different delay-Doppler domain parameters.