以定位為基礎之網路優化軟體平台研究與開發

無線網路的優化，可以使得系統業者，以最低的成本，建構出最有益的通訊網路，達到最大的覆蓋範圍及容量(Capacity)和最好的服務品質;而傳統上系統業者是以人工方式來進行優化，其結果常常不是最佳解答，且需花費許多時間及大量人力，所以紛紛轉向自動優化系統的懷抱。而本論文將延續本實驗室所設計之軟體平台，建構一套以定位為基礎之網路自動優化軟體平台，並且在優化過程中加入定位資訊，不僅可以使得優化目標可以更加明確，也可以取代傳統上，系統業者派工程師至各地量測之方案。在定位演算法方面，將引用本實驗室所發展之訊號衰減差異(Signal Attenuation Difference, SAD)方法，為軟體平台之定位演算法，而SAD之方法不需要精準地傳播模型，並且可以降低遮蔽效應對於定位準確度影響，以及不必更改硬體，即可以在目前通訊系統中執行。而在優化演算方面，將使用多目標基因演算法(Multi-Objective Genetic Algorithm, MOGA)來克服多目標之非線性優化問題並產生多個權衡優化解(multiple trade-off optimal solution)，相較於單一目標基因演算法在處理相互衝突之多目標函數而言，增加許多彈性及更有能力處理行動通訊系統優化問題。最後，我們將整合上述之定位及優化演算法於軟體平台中，且進一步利用平台分析及探討在不同適應性函數(Fitness Function)及不同定位演算法之下，對於網路優化後之效能。

關鍵字

多目標基因演算法；訊號衰減差異；柏拉圖最佳解；網路優化

並列摘要

Wireless network performance enhancement platform helps network operator that can achieve a profitable network that maximizes coverage, capacity and quality at minimal cost. Traditionally , network operator use manual planning and performance enhancement method which is a time-consuming and manpower-consuming work and rarely produces the best network configuration. Hence automated performance enhancement method becomes an alternative approach for network operator. The purpose of this thesis is to develop a location based network performance enhancement software platform. Including location information in performance enhancement process not only makes a target of optimization more clearly but also replaces the traditional optimization approach that network engineers have to do the measurement everywhere. On the locaiton algorithm , we introduce a SAD(Signal Attenuation Difference) method in our software platform. SAD method doesn’t need precise propgation model and reduces the shadowing impacts on the location accuracy. Furthermore, SAD method can be emplyed in current communication system without changing any hardware. In the case of optimization algorithm , we employs the multi-objective genetic algorithm to overcome multi-objective non-linear optimization problem and generate multiple trade-off optimal solutions . Multi-objective genetic algorithm has more ability and flexibility than single-objective genetic algorithm to deal with mobile communication system optimization problems. Finally , we integrate location algorithm and optimization algorithm given above into software platform. Moreover, we apply this platform to analyze and discuss the network performance after optimization in different fitness functions and different location algorithms.

並列關鍵字

Multi-Objective Genetic Algorithm ； Signal Attenuation Difference ； Pareto Optimal Solution ； Network Performance Enhancement

參考文獻

[1] J. M. Zagami, S. A. Parl, J. J. Bussgang, and K. D. Melillo, “Providing Universal Location Services Using a Wireless E911 Location Network,” IEEE Communications Magazine, Vol. 36, pp.66-71, April. 1998.

[2] D. Kalyanmoy, Multi-Objective Optimization using Evolutionary Algorithm, John Wiley & Sons, New York, 2001.

[4] S. Yamada, R. Enomoto, J. Takayama, and S. Ohyama, “Sensor nodes localization based on inequalities of electric field intensity,” Sensors, 2005 IEEE, 30 Oct.-3 Nov. 2005.

[5] L. M. Ni, Y. Liu, Y. C. Lau, and A. P. Patil, “LANDMARC: indoor location sensing using active RFID,” Proceedings of the First IEEE International Conference on Pervasive Computing and Communications, pp. 407-415, March. 2003.

[7] W. Foy, “Position-Location Solutions by Taylor-Series Estimation,” IEEE Transactions on Aerospace and Electronic Systems, Vol. AES-12, Issue 2, pp. 187–194, March. 1976.

國際替代計量

以定位為基礎之網路優化軟體平台研究與開發

全文下載

主題瀏覽