Scholarlink

<p>The evolution towards sixth-generation (6G) wireless networks necessitates sophisticated optimization techniques to address the unprecedented complexity of multi-band communication systems and stringent quality-of-service requirements across diverse applications. This paper presents a novel artificial intelligence-driven cross-layer protocol optimization framework specifically designed for multi-band 6G wireless communications, integrating deep reinforcement learning with adaptive network slicing to enable dynamic parameter optimization across physical, data link, network, and transport layers. The proposed hybrid CNN-LSTM architecture combines convolutional neural networks for spatial feature extraction with long short-term memory networks for temporal pattern recognition, enabling real-time prediction of optimal protocol configurations across Sub-6&#xa0;GHz, millimeter-wave, and terahertz frequency bands. The framework employs a multi-objective optimization approach that simultaneously maximizes spectral efficiency, minimizes energy consumption, reduces latency, and mitigates interference through intelligent coordination between frequency bands. Comprehensive simulation results demonstrate significant performance improvements, achieving 34.7% enhancement in spectral efficiency (18.8 bits/s/Hz), 28.3% reduction in energy consumption (95.2 Mbits/Joule), and 41.2% decrease in end-to-end latency (5.5&#xa0;ms) compared to traditional cross-layer approaches. The proposed framework successfully adapts to dynamic network conditions across urban, suburban, and rural deployment scenarios while maintaining consistent performance advantages, establishing a foundation for intelligent protocol optimization in next-generation wireless communication systems.</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

AI-driven cross-layer protocol optimization for multi-band 6G wireless communications

  • 期刊论文

  • 发表日期：

    2026年5月26日

查看全文
  • Abdulmohsen Mutairi

摘要

The evolution towards sixth-generation (6G) wireless networks necessitates sophisticated optimization techniques to address the unprecedented complexity of multi-band communication systems and stringent quality-of-service requirements across diverse applications. This paper presents a novel artificial intelligence-driven cross-layer protocol optimization framework specifically designed for multi-band 6G wireless communications, integrating deep reinforcement learning with adaptive network slicing to enable dynamic parameter optimization across physical, data link, network, and transport layers. The proposed hybrid CNN-LSTM architecture combines convolutional neural networks for spatial feature extraction with long short-term memory networks for temporal pattern recognition, enabling real-time prediction of optimal protocol configurations across Sub-6 GHz, millimeter-wave, and terahertz frequency bands. The framework employs a multi-objective optimization approach that simultaneously maximizes spectral efficiency, minimizes energy consumption, reduces latency, and mitigates interference through intelligent coordination between frequency bands. Comprehensive simulation results demonstrate significant performance improvements, achieving 34.7% enhancement in spectral efficiency (18.8 bits/s/Hz), 28.3% reduction in energy consumption (95.2 Mbits/Joule), and 41.2% decrease in end-to-end latency (5.5 ms) compared to traditional cross-layer approaches. The proposed framework successfully adapts to dynamic network conditions across urban, suburban, and rural deployment scenarios while maintaining consistent performance advantages, establishing a foundation for intelligent protocol optimization in next-generation wireless communication systems.