透過無線傳輸方式的資料量在近十年大幅增長，並且預期在下個世代 對於無線傳輸的需求將是有增無減。為了應對這樣大量增長的需求，提高 無線傳輸的效率和傳輸能力是必然的趨勢近期，Opportunistic Scheduling (OS)的概念在許多的通訊相關的研究領域被大量探討。OS是在考慮通 道品質的情況下，以機遇式方式對無線通道進行排程來改善無線傳輸時 的效率，排程器會挑選接收訊號品質相對較好的使用者優先傳輸，如此 一來系統的平均吞吐量將會改善許多。論文的主題除了將OS實現在群播 中，同時能應用於Cognitive Radio (CR)與Multimedia Broadcast multicast service Single Frequency Network (MBSFN)的環境下。探討並克服在上述 兩個情境底下，Opportunistic Multicasting (OM) 排程如何設計才能提升 整體吞吐量。首先，我們探討了Multiuser Diversity (MUD)的理論分析， 並提出統一內部及外部MUD的概念，此概念可以讓我們對系統整體吞吐量 和使用者吞吐量進行公式化，進而在設計CR的演算法時能夠量化吞吐量。 第二，我們針對不同Modulation and Coding Scheme (MCS)的調整間隔， 探討了Signal-to-interference-plus-noise Ratio (SINR)在MBSFN的分布情 形和OM的頻譜效率。運用此結果，我們設計出一低回饋、低複雜度和慢 速調整調變編碼的演算法應用在MBSFN。 綜合以上兩點，我們所設計的演算法，非常適用於CR和在MBSFN的 環境下。大量的模擬結果也顯示我們所提出的演算法比起此前的設計能提 供較佳的服務品質給用戶。;The demand for wireless data is continuously rising and is expected to increase. To cope with this growth, it is necessary to work towards higher network capacity and improved network capabilities. In the recent period, opportunistic concepts started to take hold and being part of research into communication technologies. Opportunistic Scheduling (OS) is a scheduling approach for wireless channels, where channel quality is considered, when making scheduling decisions. The scheduler transmits to users with relatively better signal quality, so the average amount of transmitted data can be increased. The focus of this thesis will be on OS as a method for increasing the transmission e ciency and its implementation in wireless networks. The main research topic of the dissertation is thus multicasting with OS, i.e. Opportunistic Multicasting (OM), and its adaptation to Cognitive Radio (CR) and Multimedia Broadcast multicast service Single Frequency Network (MBSFN) networks. More speci cally, how it ts into the these types of networks and what kind of obstacles need to be tackled, before OM can be successfully deployed. First we study the theory of Multiuser Diversity (MUD) and provide a novel concept unifying internal and external MUD, enabling us to formulate the system sum and user throughput. The resulting CR algorithm design approaches the derived throughput advantage. Second, we look at Signal-to-interference-plus-noise Ratio (SINR) distribution in MBSFN networks and Spectrum E ciency (SE) of OM under extended Modulation and Coding Scheme (MCS) adaptation interval. Using the results, we design a low feedback, low complexity, and slow MCS change based algorithm suited for MBSFN. The resulting algorithm designs are well suited to the studied CR and MBSFN networks. Extensive simulations show, that the presented algorithm designs perform better than previous designs, enabling better quality service to users.
