近年來 LTE 用戶規模快速增加，加上行動影音服務的蓬勃發展，促使數據流量呈爆炸性成長，為了因應使用者大量的傳輸需求，3GPP於2008年提出LTE-Advanced ，使用載波聚合技術將多個LTE載波單元(Component Carrier，CC)聚合在一起，以實現最大的傳輸頻寬，有效提高傳輸速率。 雖然載波聚合能有效的提高傳輸速率，但現有授權頻譜的聚合使用仍然無法應付逐漸增長的傳輸量，因此 3GPP 便提出將載波聚合技術拓展至未授權頻譜 ( Unlicensed Band )，即為 LTE-U。本篇論文即是研究在LTE-U環境下，如何有效且彈性的分配授權頻譜及未授權頻譜資源，我們先根據 LTE或 LTE-U 用戶給予不同權重大小接著做初步排程，藉由此機制讓 LTE 用戶在初步排程時的競爭更有優勢，然後我們確認 GBR 用戶是否滿足傳輸需求，如果部分用戶未滿足時，則對特定用戶進行有條件的插隊與佔用程序機制，此佔用程序機制考慮到用戶類型、訊務類型以及佔用期間的吞吐量數值，並設置機制保護 Non-GBR 用戶以避免無法獲得傳輸的情況，我們期望這個機制允許用戶將損失降到最低，以換取用戶最大滿意度。 最後模擬顯示提出的演算法能夠彈性的分配資源，保障 LTE 用戶也有足夠資源，且相較於其他演算法更能有效提升整體用戶滿意度。 ;In recent years, LTE users have increased rapidly and the vigorous development of mobile audio & video services to promote explosive growth in network data traffic. In order to meet the user requirements, 3GPP proposed LTE-Advanced in 2008, using carrier aggregation technology to aggregate multiple LTE component carrier together to achieve the maximum transmission bandwidth, effectively improve the downlink transmission data rate. Although carrier aggregation can effectively increase the transmission rate, the use of existing licensed spectrum is still to cope with the increase in the amount of transmission. Therefore, the 3GPP has proposed to expand the carrier aggregation spectrum to unlicensed spectrum, LTE-U. This thesis is to study how effective and flexible allocation of licensed spectrum and unlicensed spectrum resources in LTE-U environment. In order to make the LTE user more competitive we assign LTE and LTE-U users different weights and then test the initial schedule. Then we will determine whether the GBR (Guaranteed Bit Rate) user’s data rates satisfied their requirements. If the rates are unsatisfactory for some GBR users, we can determine the specific user conditional queue and occupation schedule. Next our proposed scheduling process will prioritize the user type, the type of service, and the throughput value during occupancy. This may result in lower transmission rates for non-GBR users, depending on the needs of the situation. Finally, the simulation shows that the proposed algorithm can allocate resources flexibly to ensure that LTE users also have sufficient resources and are more effective than other algorithms to improve user satisfaction.
