在超密度小型基地台網路中兼顧用戶設備能源效率及網路負載之移動性管理

Abstract

超密度網路由於具延展性且佈建成本低廉，被視為達到高網路容量極具潛力的解決方案。然而，微型基地台的密化將降低細胞網路移動效能之強健性。當用戶設備以高速移動時，它會有不佳的移動性效能與高能量耗損，特別是它在長間斷接收周期中時。並且在這種場景下，傳呼和追蹤區域更新所導致對核心網路的負載將更高。在這篇論文中，我們探討如何智能地做移動性量測以及規劃追蹤區域表單。我們首先推導在兩種移動模型下，用戶設備從兩個同心圓的內圓走到外圓所花的時間之機率分布。我們分析無線電鏈路故障率及換手的時間點，並提出一個最小化功率消耗的量測頻率調整機制。我們也提出一個目標式以最小化網路信息負載，並且同時避免大量突發追蹤區域更新。我們提出了一個基於模擬退火的演算法、與一個運算快速的啟發式演算法，來為每個用戶移動設備選擇追蹤區域表單。我們的模擬結果顯示在超密度網路中，我們的機制及演算法能夠節省許多能量，並且達成低網路信息負載及更少的突發更新。

Ultra-dense network is considered a promising solution for high network capacity. It is favored for its scalability and cost effectiveness. However, small cell densification reduces the mobility robustness. When a user equipment is at high speeds, it may suffer from poor mobility performance and high power consumption, especially when it is in a long discontinuous reception cycle. In addition, the signaling load to the core network resulted from the paging and the tracking area update will be exacerbated in such scenario. In this thesis, we investigate how to perform mobility measurements and tracking area list planning intelligently. We first derive the probability distribution of the time it takes a user equipment to move from the inner to the outer co-centric circles for two mobility models. We analyze the radio link failure rate and the timing of handover, and then propose a mechanism to adjust measurement frequency to minimize the power consumption. We also formulate an objective to minimize the network signaling load with consideration of avoiding tracking area update signaling storm. We propose an algorithm to choose tracking area list for each user equipment based on simulated annealing, as well as a computationally efficient heuristic algorithm. Our simulation results demonstrate our mechanism and algorithms can save considerable amount of energy and achieve low network signaling load with few bursting updates in ultra-dense networks.