任意網路拓樸下小型基地台佈建之最佳化

Abstract

異質網路下小型基地台的佈建近年來被視為提升覆蓋範圍與通道容量的核心技術，有別於傳統根據使用者即時的需求來佈建小型基地台，越來越多的研究提出更具管理性的佈建模型來權衡基地台性能與其佈建成本。由於小型基地台可以被佈建在屋頂或街燈上來服務鄰近的戶外使用者，以往相關文獻所提出的室內佈建策略將不再適用。此外，過去相關文獻對小型基地台所假設的隨機分佈模型也無法適用於任意網路拓樸。有鑑於此，本論文分別探討了在共享和獨立資源分配模型下，小型基地台在開放模式與封閉模式的最佳佈建策略。為了達到這個目的，我們透過最佳化小型基地台佈建位置與操作參數，來最大化不同服務質量需求的使用者數量。由於這個最佳化問題屬於混合整數及實數之非線性規劃，為了簡化其計算的複雜度，我們提出了一個可以隨意控制計算時間的演算法，把原本問題拆解為「群組形成」與「資源管理」的兩個子問題：群組形成子問題著重在找尋最佳的佈建地點以及服務範圍；而資源管理子問題則著重在控制傳輸功率以及合理地分配無線資源使每個使用者都能滿足其個別的服務質量需求。此外，本論文也將此佈建策略延伸擴展到另一個後最佳化問題，意即當所有佈建都已完成後，倘若使用者服務質量需求改變，抑或者有新的使用者加入現有網路，我們如何最小化新佈建的小型基地台來因應這個變化。最後，模擬的結果顯示本論文提出的佈建策略相較於其他佈建策略，更能有效地在計算複雜度與最佳化程度之間取得一個完美平衡。

Heterogeneous network with small cells has recently been regarded as a promising scenario for enhancing macrocell coverage and/or capacity in LTE-Advanced systems. While deployment of small cells has typically followed the bottom-up paradigm driven by the ad hoc demand of users, more and more studies have prompted a move towards a more managed deployment model for better tradeoff between performance and cost. Unlike related work that assumes a stochastic distribution model for small cells, in this thesis we consider the deployment problem for arbitrary wireless networks under different network scenarios, including shared and dedicated resource models as well as open and closed access modes for small cells. To proceed, we formulate an optimization problem for small cell deployment in an arbitrary network that involves determination of deployment locations and operation parameters to maximize the supported number of customers with QoS constraints. Since the formulated problem belongs to mixed-integer non-linear programming (MINLP), we propose an anytime algorithm that transforms the joint problem into a cluster formation sub-problem (involving location selection and cell coverage) and a resource management sub-problem (involving power control and resource allocation) for effectively solving all optimization variables in an iterative fashion. Finally, the proposed approach is extended for solving the post-optimization problem where QoS constraints and/or number of customers are changed after initial network planning. Compared with other approaches for small cell deployment, evaluation results show that the proposed algorithm can effectively solve the target problem while striking a better performance tradeoff between computation complexity and solution quality.