- 學位論文
在IEEE 802.11be WLAN中使用強化式學習增強協調式空間複用以提高效能
Increasing Coordinated Spatial Reuse in IEEE 802.11be WLANs for Enhancing Throughput: A Reinforcement Learning Scheme
摘要
隨著IEEE 802.11be標準（Wi-Fi 7）的推出，無線通訊技術實現了資料傳輸速率和頻譜利用率的顯著提高。特別地，協調空間重用（C-SR）技術對於提升多接入點（AP）環境中的網路效能具有關鍵作用。然而，優化C-SR策略需要精確調整傳輸功率和選擇合適的調製編碼方案（MCS），這對傳統網路管理構成挑戰。本研究提出了一種基於強化學習的方法，動態調整傳輸功率和MCS選擇，以提高網路的信號品質和輸送量。通過模擬實驗，我們驗證了該方法能顯著優化高密度無線區域網路(Wireless LAN, WLAN)環境下的網路效能，相較於傳統策略，顯著提高了網路的總輸送量，並證明瞭強化學習在動態網路環境中調整功率和MCS的有效性。本研究為WLAN的智慧管理和自動化提供了有效的解決方案，推動了無線通訊技術的發展，並為實現更高效、可靠的無線連接提供了實際參考。
並列摘要
With the launch of the IEEE 802.11be standard (Wi-Fi 7), wireless communication technology has achieved significant improvements in data transmission rates and spectrum utilization. Specifically, Coordinated Spatial Reuse (C-SR) technology plays a key role in enhancing network efficiency in multi-access point (AP) environments. However, optimizing C-SR strategies requires precise adjustment of transmission power and selection of appropriate Modulation and Coding Schemes (MCS), which pose challenges for traditional network management. This study proposes a reinforcement learning-based method to dynamically adjust transmission power and MCS selection to improve network signal quality and throughput. Through simulation experiments, we validated that this method can significantly optimize network performance in high-density Wireless Lan environments, substantially increasing the total network throughput compared to traditional strategies, and proving the effectiveness of reinforcement learning in adjusting power and MCS in dynamic network environments. This research provides effective solutions for the intelligent management and automation of WLAN, advancing the development of wireless communication technology, and offering practical references for achieving more efficient and reliable wireless connections.