arXiv:2611.07311v1 Announce Type: new Abstract: Accurate and efficient channel modeling is critical for evaluating the sixth-generation (6G) networks, while dynamic ray-tracing (DRT) remains bottlenecked by its reliance on channel coherence time. To address this limitation, this paper proposes a novel bounding volume hierarchy-accelerated DRT (BVH-DRT), which introduces a mechanism shift from a time-driven update in DRT to a spatial-driven update. The proposed channel model clusters complex scatterers using the BVH method as pre-processing to establish a geometric coherence distance guided by transceiver kinematics, so that the ray update intervals in BVH-DRT can be autonomously adjusted to reduce the complexity. Furthermore, the non-stationary channel impulse responses are generated to evaluate delay-Doppler (DD) domain waveforms, including orthogonal time frequency space (OTFS) and orthogonal delay-Doppler division multiplexing (ODDM). Validations in vehicle-to-vehicle (V2V) scenarios demonstrate that BVH-DRT achieves spatial-temporal consistency with measurements and commercial software while significantly enhancing computational efficiency. Additionally, bit error rate (BER) evaluation highlights the severe impact of high-dynamic channels on waveform performance. The proposed framework is important for designing reliable communications and robust sensing in 6G networks.