Network Calculus‐Based <scp>QoE</scp> Optimization for Cross‐Cultural Real‐Time Video Communication

ABSTRACT With the acceleration of globalization, immersive virtual interactive art has emerged as a critical medium for cross‐cultural communication, yet its effectiveness is severely hindered by the stochastic nature of international network transmission. This paper proposes a novel Network Calculus (NC) based framework to optimize the Quality of Experience (QoE) in real‐time cross‐cultural video communication. Unlike traditional heuristic or machine learning‐based approaches that provide probabilistic guarantees, our method utilizes algebra to derive deterministic upper bounds for end‐to‐end latency and jitter, ensuring the maintenance of environmental realism and emotional resonance across heterogeneous borders. We establish a multi‐fractal arrival envelope to model complex cultural content and apply the pay‐burst‐only‐once (PBOO) principle to minimize delay accumulation in tandem trans‐border nodes. Experimental results using the Questset and SalViT360 datasets demonstrate that the proposed method reduces the delay violation probability to less than 0.04% and significantly enhances the Subjective Perceptual Fidelity compared to eight state‐of‐the‐art methods. This research provides a robust mathematical foundation for the digital preservation and global dissemination of cultural heritage.