行動自組織網路的頻譜與功率分配機制

Abstract

隨著科技的演進，無線通訊裝置的數量成長，以及更多的高頻寬服務，都將對於行動無線網路造成許多的問題及挑戰，包括了無線使用者人數的成長、高頻寬的需求、功率的消耗以及網路的動態負載量。為了要能夠有效的處理上述問題，因此嘗試使用雲端的特性來解決，而我們所提出的系統架構C2RAN便是建構於雲端上，可藉由自組織網路 (self-organization network, SON) 技術的中控式資源分配管理，遍佈的無線接取點 (access point, AP)， 以及對於移動使用者的排程，將更有效率的動態使用頻譜，進行服務。在本篇論文中，為了在服務越多使用者的前提之下使用最少的功率，我們針對資源分配提出了兩種演算法，並探討如何對於移動性做演算法有效的改進；最後以台北市為一個大尺度的模擬，根據台北市的統計數據，模擬真實環境下的AP布建方式和利用移動模型模擬使用者真實移動的狀況，去驗證C2RAN確實能夠服務大尺度的台北市。

With the rise of the number of smart handheld devices, the demand of wireless communication services and higher bandwidths also increase accordingly, and these demands will cause many problems and challenge issues for mobile wireless networks, including the wireless spectrum resource sharing as well as the power consumption. In order to effectively deal with the above problems, the consensus so far is to distribute numbers of wireless access points (APs) everywhere, which yields more challenge issues of interference control and resource management among APs. To this end, we proposed a cognitive cloud radio access network (C2RAN) to centrally and dynamically allocate the channel and power resources to the mobile users (MUs) and APs. Following the concepts of self-organizing network (self-organization network, SON), we use a centralized resource allocation and user scheduling to increase the system spectrum reuse. In this thesis, in order to serve as many MUs as possible with consuming the minimal power to meet each of their requirements, we propose two resource allocation algorithms, and explore how to improve the efficiency for mobility. Finally, we provide a large-scale simulation with taking Taipei city as an example. In order to justify the feasibility of C2RAN in such a scale, we simulate the real environment, including the location of APs and the mobility model of the mobile users which are all simulated according to the statistics of Taipei.