AI-Driven Resource Allocation and Scheduling in 6G Networks: A Comprehensive Review of Deep and Reinforcement Learning Approaches

The emergence of the sixth-generation (6G) wireless communication networks has put unprecedented demands for ultra-low latency, massive connections, high spectral efficiency, and intelligent adaptability. Due to the increased complexity of the system and the dynamic environment, traditional model-based resource management methods are challenging in meeting these constantly changing demands. This paper comprehensively investigates AI-driven resource allocation and scheduling methods in 6G networks, focusing on deep learning (DL) and reinforcement learning (RL). We explored how to apply deep learning models such as CNN and RNN to traffic prediction, channel estimation, and intelligent spectrum allocation. Meanwhile, reinforcement learning techniques - including Q-learning and deep Q-networks (DQN) - have demonstrated robust adaptive power control, dynamic spectrum access, and distributed multi-agent coordination capabilities. This review compares the advantages and limitations of deep learning and reinforcement learning through a detailed analysis of the latest applications and frameworks. Their complementary roles are achieving intelligent, flexible, and efficient resource management. Finally, we highlighted future research directions such as hybrid learning models, decentralized architectures, and energy-aware optimization, which are expected to support sustainable and scalable AI-driven 6G systems.

Preparing paper insights from the available abstract and paper details.

Authors

Yu Chen

Published

Mar 10, 2026

Journal or venue

Advances in Engineering Technology Research

DOI

10.56028/aetr.16.1.207.2026

Paper type

journal article

Source

Crossref