5G毫米波網路中用戶連線與資源配置演算法之設計

Abstract

5G毫米波網絡可透過使用波束成形(beamforming)技術、與小細胞接入點(small cell access points, AP)的密集部署，以提供Gbps等級的速率。在本論文中，我們考慮用戶連線與通道資源分配的問題。該問題可以表示為整數非線性最佳化。為了解決該問題，我們首先利用對偶分解(dual decomposition)法來獲得基準場景的集中式(centralized)演算法。在基準場景中，並沒有考慮用戶端需求。接著，我們將用戶本身的需求分別考慮在限制中或作為目標中的權重，然後提出遞迴式與分散式(distributed)演算法。數值結果顯示我們提出的兼具用戶需求考量的演算法在吞吐量方面可明顯地優於基準場景之演算法。

A 5G millimeter wave network is able to deliver speeds of Gbps through the use of beamforming and the dense deployment of small cell access points (APs). In this thesis, we consider the problem of the associations of clients and allocations of channel resources. The problem can thus be formulated into an integer nonlinear optimization. To solve the problem, we first utilize a dual decomposition method to obtain a centralized algorithm for a baseline scenario. In the baseline scenario, no client demands are considered. Next, demands of clients are taken into account either in the constraints or as weights in the objective, and a distributed algorithm is then proposed. Numerical results demonstrate that our proposed algorithm for scenarios with client demands can substantially outperform the one for scenarios without them.