都會區行動網路中基於移動行為預測之換手決策

隨著無線行動通訊網路日益成長以及智慧型手機等裝置的蓬勃發展，人們對於頻寬和網路覆蓋率的要求越來越高。因此，較小型的基地台如Microcell、Picocell、Femtocell等相繼出現以增加網路容量及覆蓋率。在都會區，由於網路使用者眾多，對於頻寬有著更高的要求，所以基地台也佈建較為密集。但是小基地台的增加也造成換手次數愈趨頻繁。過多的換手可能造成通訊延遲、中斷等問題，同時也會增加對基地台的負擔。因此，在無線行動通訊網路中如何減少不必要的換手次數是相當重要的課題。以往的傳統換手決策總是選擇訊號強度(Received Signal Strength, RSS)最強的基地台當作是換手目標基地台(target cell)，卻沒有考慮到使用者的移動性。所以容易產生不必要的換手。本論文利用人們的移動行為會因為地形因素有所限制的特性，提出了基於使用者移動行為以預測選擇目標基地台(Mobility Prediction based Handover Decision, MPHD)的機制來避免多餘的換手。本研究於考慮遮蔽效應(shadow fading)及路徑損失(Pathloss)的環境中，使用Mobility Markov Chain(MMC)為主要預測工具，利用目前及之前的服務基地台(serving cell)組合來預測未來所會選擇的目標基地台(target cell)。在本論文中所使用之移動模型為韓國某大學校園內學生之真實移動路徑。本研究所提出之演算法無需全球定位系統(Global Positioning System, GPS)等裝置輔助，也不必知道User Equipment(UE)的移動速度即可使用。最後與傳統換手決策比較後發現MPHD的確可以在訊雜比(Signal to Interference-plus-Noise Ratio, SINR)的表現不會比傳統換手決策差的情況下減少換手次數。

關鍵字

換手決策；行動預測； Mobility Markov Chain ；真實移動路徑；減少多餘換手

並列摘要

Recently, cellular network is a rapidly growing segment of communications market. People spend more and more time in using intelligent mobile phones. In order to provide larger network capacity and coverage especially in dense urban area with high population density where many people may simultaneously use the network, cellular operators deploy a lot of small base stations like microcells, picocells, and femtocells. Installation of small base stations not only increases network coverage but also handover frequency. Frequent handovers may cause the transmission delay longer and service interruption, and also increase loading of base stations. Therefore, how to reduce unnecessary handovers becomes an important issue in cellular networks. Regardless of user equipment’s mobility, conventional handover decision always chooses the cell with strongest signal strength as the target cell. Thus, the conventional handover decision in dense urban area may easily cause unnecessary handovers. This study proposed a predictive mechanism: Mobility Prediction based Handover Decision (MPHD), which according to users’ moving behavior can choose an appropriate base station as the target base station to reduce unnecessary handovers. The proposed scheme predicts the target cell that the UE may choose in the future with the serving cell and several previous serving cells by using the Mobility Markov Chain. We assessed our scheme with mobility models from real moving logs. These logs were taken by college students in a campus in Korea. The MPHD needs neither positioning equipment like global positioning system (GPS) nor knowing the speed of the UE. According to the simulation result, our proposed mechanism indeed reduces unnecessary handovers. The Best of all, MPHD mechanism doesn’t lead to SINR degradation comparing to conventional handover decision.

並列關鍵字

handover decision ； mobility prediction ； Mobility Markov Chain ； real moving log ； reducing unnecessary handover

參考文獻

[7] S. J. Koh, “Multicast Handover Agents for Fast Handover in Wireless Multicast Networks,” IEEE Communications Letters, vol. 14, no. 7, pp. 676-678, July 2010.

[9] T. Guo, A. Quddus, and R. Tafazolli, “Seamless Handover for LTE Macro-Femto Networks Based on Reactive Data Bicasting,“ IEEE Communications Letters, no. 99, pp. 1-4, Nov. 2012.

[10] L. Xie, Q. Li, W. Mao, J. Wu, and D. Chen, “Achieving Efficiency and Fairness for Association Control in Vehicular Network,“ 17th IEEE International Conference on Network Protocols, pp. 324-333, Oct. 2009.

[11] Y. Zhao, W. Li, J. Hong, Z. Li, S. Lu, and D. Chen, “On handoff Minimization in Wireless Networks: From a Navigation Perspective,“ IEEE

Wireless Communications and Networking Conference, pp. 1-6, Apr. 2010.

國際替代計量

都會區行動網路中基於移動行為預測之換手決策

未授權

主題瀏覽