Scholarlink

<p>This paper proposes an adaptive hybrid non-orthogonal multiple access (AH-NOMA) framework for fifth generation (5G) networks, leveraging a Deep Q-Network (DQN) to dynamically manage resource allocation under user mobility. Conventional hybrid NOMA/OMA schemes rely on static switching rules, leading to suboptimal performance in dynamic environments. The proposed framework formulates the mode selection, user pairing, and power allocation problem as a Markov Decision Process (MDP). Its novelty lies in a mobility-aware state design that incorporates real-time channel conditions and user velocity, and a multi-objective reward function that jointly optimizes spectral efficiency, latency, and fairness. Simulations demonstrate that the DQN-driven AH-NOMA framework achieves a 50% improvement in spectral efficiency and a 60% reduction in latency compared to static schemes, while also enhancing fairness and reducing successive interference cancellation (SIC) complexity. The results confirm the framework’s viability for intelligent resource management in mobile 5G deployments.</p>

错误：搜索内容不能为空，请输入英文关键词
错误：关键词超出字数限制，请精简
高级检索

RL-DQN-driven adaptive hybrid NOMA for 5G networks: dynamic mode switching and mobility-aware optimization

  • 期刊论文

  • 发表日期：

    2025年11月21日

查看全文
  • R. Chandrasekhar,
  • Poonam Singh

摘要

This paper proposes an adaptive hybrid non-orthogonal multiple access (AH-NOMA) framework for fifth generation (5G) networks, leveraging a Deep Q-Network (DQN) to dynamically manage resource allocation under user mobility. Conventional hybrid NOMA/OMA schemes rely on static switching rules, leading to suboptimal performance in dynamic environments. The proposed framework formulates the mode selection, user pairing, and power allocation problem as a Markov Decision Process (MDP). Its novelty lies in a mobility-aware state design that incorporates real-time channel conditions and user velocity, and a multi-objective reward function that jointly optimizes spectral efficiency, latency, and fairness. Simulations demonstrate that the DQN-driven AH-NOMA framework achieves a 50% improvement in spectral efficiency and a 60% reduction in latency compared to static schemes, while also enhancing fairness and reducing successive interference cancellation (SIC) complexity. The results confirm the framework’s viability for intelligent resource management in mobile 5G deployments.