In order to facilitate intelligent, secure, and ultra-low-latency communications, the proposed Cognitive 6G framework combines quantum computing, mobile edge computing, and artificial intelligence (AI) into a single multilayer architecture. While the MEC layer uses lightweight AI models for local inference, caching, and real-time control, heterogeneous IoT nodes provide multimodal data at the device layer. Large-scale model training and collaboration with MEC nodes sustain global intelligence at the cloud AI layer. Complex optimization tasks like resource allocation, beamforming, and quantum key distribution-based security are accelerated by a dedicated quantum layer. Real-time cognition, quantum- safe transmission, and context-aware network adaptation are made possible by this close integration. The AI6GQoE dataset is subjected to machine learning models in order to assess user-centric performance. Random Forest regression yields an MAE of 0. 32, RMSE of 0. 47, and a R2 of 0. 81. The results show that AI-driven learning has the potential to improve user experience in emerging 6G scenarios, such as Open RAN and blockchain-enabled networks, and they also indicate good QoE prediction capability.
Bennet et al. (2026) studied this question.