黃政吉王嘉斌Huang, Jeng-JiWang, Chia-Pin陳柏丞Chen, Po-Chen2024-12-172024-07-312024https://etds.lib.ntnu.edu.tw/thesis/detail/8c9b4cf5b7220ad491810383da4f8a6a/http://rportal.lib.ntnu.edu.tw/handle/20.500.12235/1229545G毫米波網絡(millimeter Wave, mmWave)是一種高頻信號，波長在毫米範圍內。由於毫米波的波長極短且穿透能力差，並且其有效範圍有限，本論文考慮到這些特性，將其與最大最小公平資源分配相結合。類似於無線局域網（Wireless LANs），接入點（AP）之間的流量負載分佈通常是不均衡的，導致用戶之間的頻寬分配不公平。通過最大最小資源分配，可以顯著減少負載不均衡和隨之而來的不公平頻寬分配。在本論文中，我們提出了一種有效的解決方案來確定用戶與接入點的連接，以實現最大最小公平的頻寬分配。我們展示了公平性與負載平衡之間的密切關係，這使我們能夠利用負載平衡技術來實現最佳的最大最小公平頻寬分配。我們的模擬結果表明，所提出的部份連線演算法接近於實現最佳負載平衡。最後，我們將我們的方法與比例公平資源分配進行了比較，並探討了兩者之間的差異及其優缺點。The 5G millimeter wave network is a type of high-frequency signal with wavelengths in the millimeter range. Due to the extremely short wavelength and poor penetration capability of millimeter waves, as well as their limited effective range, this paper considers these characteristics and integrates them with max-min fairness resource allocation. Similar to wireless LANs, the distribution of traffic load among access points (APs) is typically uneven, resulting in unfair bandwidth allocation among users. Through max-min resource allocation, significant reductions in both load imbalance and consequent unfair bandwidth distribution can be achieved. In this paper, we propose an effective solution for determining user-AP connections to achieve max-min fair bandwidth allocation. We demonstrate the close relationship between fairness and load balance, which enables us to utilize load balancing techniques in order to attain optimal max-min fair bandwidth allocation. Our simulation results indicate that the proposed fractional-association close to algorithm achieves optimal load balancing. Finally, we compare our approach with the related work proportional fairness resource allocation, and we examine the differences and discuss merits between the two approaches.集中式演算法最大最小公平性單行法5G毫米波網路拉格朗日對偶分解Centralized AlgorithmMax-Min Fairnesssimplex MethodMillimeter Wave NetworkLagrange Dual Decomposition學術論文