載波聚合(CA)是LTE-Advanced系統相當重要的技術,透過CA將頻譜上可利用的載波聚合一起以形成更大頻寬給特定使用者使用來提升傳輸速率。LTE-Advanced系統的無線電資源管理(RRM)由成員載波選擇及封包排程兩種資源分配組成,本論文針對非連續CA提出一套資源分配演算法,並建立系統級模擬平台驗證演算法的效能改善。 因為不同頻帶的通道特性不一樣導致使用者能聚合的載波數不相同,本論文的成員載波選擇方式是由使用者回報的SINR值去設定載波分配的臨界值。而為了提升通道差之使用者的傳輸量,我們利用使用者分組的方式將聚合相同數量載波的使用者視為一個組群,每個群組有不同的資源分配權重值,藉由這樣的方式來改變不同群組得到資源排程的機會。 由系統級模擬平台驗證載波聚合能有效提升系統吞吐量,而採用SINR臨界值的方法決定各載波能服務的使用者,相較於參考文獻採用的路徑損耗較能提升系統平均吞吐量約0.065Mbps。另外不同群組的傳輸量也得以藉由使用者分組的方式獲得改善,在最佳值可提升約8%的排程機會,平均傳輸量約0.1124 Mbps,雖然會造成系統吞吐量的降低。但是本論文所提出的資源分配演算法,對於通道佳的使用者可以利用CA提升傳輸量,而通道差的使用者也能藉由使用者分組的方式改善傳輸量,因此可以有效平衡使用者之間的傳輸量。
Carrier aggregation (CA) is an emerging technique in LTE-Advanced system, it can aggregate many available carriers to form greater bandwidth and increase the data rate for certain user. The radio resource management (RRM) in LTE-Advanced system is composed of component carrier selection and packet scheduling. This paper focuses on non-continuous CA case and presents a resource allocation algorithm. The improvement of performance is demonstrated by system level simulation. Since channel properties in different frequency bands is various; resulting in the number of carriers which users can be scheduling on is different. In this paper, the component carrier selection method sets the carrier allocation threshold value according to the report of SINR of users. In order to increase the data rate of users who have bad channel, user grouping method is used to set users that aggregate same amount carriers as group, each group has different weighting of obtaining scheduled resources. The developed system level simulation platform demonstrates that the performance of carrier aggregation can effectively increase the system throughput. Comparing with the path loss reference method, the SINR threshold method can improve system average throughput by 0.065Mbps. Besides, the user grouping method can increase 8% of scheduling change and 0.1124 Mbps of average user date rate at optimal value, although it will decrease the system throughput. As the results shown, the proposed resource allocation algorithm can effectively balance users’ data rate for users who have good channel can use CA to increase data rate, while users who have bad channel can use user grouping to improve data rate.