考慮使用者移動行為之無線網路需求規劃及資源分配：以RNC配置為例

Abstract

隨著智慧型手機與穿戴式電子產品裝置的發展日新月異，行動通訊服務產業也跟著興起，使用者對無線網路的需求因而與日俱增，如何滿足使用者網路及服務產業的發展需求，同時可以使網路環境的控管更有效率是本研究的主軸。 無線網路需求量的不確定性除了來自時間與地區性的差異，也受到使用者移動性的影響，因此難以做出準確預測，這也是進行網路資源規劃策略上的難題；但有近期文獻指出使用者的移動性在一定的範圍下是可以被預測的，基此，本研究利用使用者的移動行為建立需求模式，並依所建立之需求模式進行無線網路資源規劃策略。 本研究利用統計探勘技巧，從資料中建立模式描述用戶移動性，我們以1小時為單位為例描述所有使用者一天中的網路需求，並利用跨越不同時段的使用者以建立轉移矩陣模型描述用戶的移動模式。再依據不同的指標與相關變數等性質及貪婪式、階層式、啟發式與基因演算法等的最佳化方法，發展多個分群與優化演算法，以進行需求規劃策略達到無線網路資源的優化。 研究最後以改善無線網路控制中心(RNC)管理基地台模式為例，希望達到提升不同管理中心的網路服務率與降低無線網路控制中心間的轉移率，該方法亦可應用於未來基地台的配置；本研究亦評估各個演算法之規劃品質及其計算效率，並比較現有的管理方式，以驗證了所提出方法的效率。

With the booming use of smart phones and wearables, the mobile service industry is also in a thriving development pace. It also causes mobile users’ increasing network demand. Our question is how to satisfy their network demand, and improve the network control and planning efficiency at the same time. The uncertainty of wireless network demand is not only from the spatial and temporal variations, but also effected by user mobility behavior. The problem is how to manage and plan the network capacity to satisfy the highly volatile demand. Recent research has shown that the user mobility can be predicted to a certain degree of accuracy. For this reason, we propose to build a model of user transition pattern to describe the user spatial and temporal mobility behavior. With the model built, the network capacity planning is optimized accordingly. This research uses statistical data mining techniques to construct the transition matrix to describe the users’ aggregated mobility behavior. The wireless network demand series is collected from all users’ daily network activity in hourly basis. The transition count between each region is also collected by selecting the user activities observed acrossing two consecutive time intervals, based on which the transition matrix is estimated. We then develop heuristic aggregation strategies based on clustering algorithm and optimization techniques, such as hierarchical clustering, k-means, greedy and genetic algorithms according to performance surrogates and objective functions. These strategies are then used to improve the efficiency of wireless network resource planning. Last, we use the example of base station (BS) control by the radio network control (RNC) to demonstrate the planning methods proposed by this research though the proposed method can be also applied in small cells allocation or fog computing resource allocation problems. The planning quality and the computing efficiency of the algorithms are also evaluated by comparing the performance among the RNC allocations planned by the proposed methods and the current allocation plan.