透過您的圖書館登入
IP:3.138.123.190
  • 學位論文

Wi-Fi無線基地台配置最佳化-以基因演算法

The Optimal Deployment of Wi-Fi Wireless Access Points Using the Genetic Algorithm

指導教授 : 王貞淑
若您是本文的作者,可授權文章由華藝線上圖書館中協助推廣。

摘要


Wi-Fi無線基地台配置的最大目標就是能讓使用者獲得無所不在的通訊品質,因此如何配置無線基地台才能在有限的預算內讓訊號的覆蓋率、預算滿足率、動態流量滿足率、干擾率和負載平衡指數達到一定的水準,便成為了重要的議題。為解決此一多目標決策問題,本研究彙整了過去學者們在Wi-Fi無線基地台配置方面的文獻,以基因演算法為基礎,提出一個Wi-Fi無線基地台配置的模型,以覆蓋率、預算滿足率、動態流量滿足率、干擾率和負載平衡指數為決策目標並建立數學模型,同時考量實務情況中有不同無線基地台類型、價錢,以及同一地點設有多個無線基地台的狀況,讓此模型能更貼近實際需求,並且為了驗證它的可行性和實用性,以校園大小為參考的需求面積,針對預算、決策權重、使用者動態流量需求分佈、負載平衡指數、干擾率、大範圍無線基地台分佈、基因參數、已設置無線基地台和不可設置無線基地台地點等不同狀況,設計了九個相關實驗,最後透過實驗,驗證本研究的實驗結果能隨著決策者的要求、使用者的需求和環境狀況,計算出讓覆蓋率、預算滿足率、動態流量滿足率、干擾率和負載平衡指數都能兼顧的解決方案,為決策者提出良好的Wi-Fi 無線基地台配置建議。

並列摘要


The purpose of Wi-Fi access point deployment is for users to receive ubiquitous communication quality. Therefore, developing methods for APs to achieve specific levels of coverage rate, budget fulfill rate, and capacity fulfill rate, interference rate, and load-balancing index within a limited budget is a vital issue. To solve such multiple objective decision-making problems, the study summarized past scholar literatures on the Wi-Fi access point deployment and proposed a model of Wi-Fi access point deployment based on generic algorithm, using coverage rate, budget fulfill rate, dynamic capacity fulfill rate, interference rate, and load-balancing index, as the decision objectives. The study further establishes a mathematical model with concurrent consideration of the different access point types, price, and deployment of multiple access points in one location in practice, so to make this model fit more with practical demand. To validate its feasibility and practicability, the study applies the campus size as reference demand for space to the different situations on budgeting, decision weight, user dynamic capacity fulfill rate, load-balancing index, interference rate, large-area access point deployment, genetic parameters, the locations already set up with access points and locations that do not allow deployment of access points by designing nine relevant experiments. Finally, the study validates the results through experiments by calculating a solution that can meet decision-maker demand, user demand and environmental conditions for coverage rate, budget fulfill rate, dynamic capacity fulfill rate, interference rate, and load-balancing index, so that the decision makers will be proposed of a good recommendation for Wi-Fi access point deployment.

參考文獻


[11] 張仕莛,Wi-Fi 自動測試環境之系統架構,大同大學資訊經營研究所,碩士論文,2010年。
[12] 詹凱翔,運用平均變異數及基因演算法來建構最佳基金投資組合權重之研究,國立成功大學財務金融研究所,碩士論文,2008年。
[16] 鐘于婷,應用於路途中資訊內容傳送之無線通訊網路評選,國立成功大學電信管理研究所,碩士論文,2004年。
[17] 謝欣宏,台鐵司機員排班與輪班問題之研究-以基因演算法求解,國立成功大學交通管理科學研究所,碩士論文,2002年。
[18] Back, T., Hammel, U. &Schwefel, H.-P., “Evolutionary computation: comments on the history and current state,” IEEE Transactions on Evolutionary Computation, Vol.1, No.1, pp.3-17, 1997.

延伸閱讀