本論文研究5G無線通訊技術中的一個領域：下行鏈路(Downlink)的資源分配和管理，在5G通訊系統的下行鏈路中，有效地處理多個Bandwidth Parts(簡稱BWP)以滿足不同應用場景的需求，是本研究的主要目標。在5G下行鏈路中，BWPs通常根據不同的Numerology(也表示為μ)來定義，包括多樣的子載波間距和時間域配置。本論文提出了一種稱為BWP Segmentation的方法，這種方法透過結合基於不同Numerology值的BWPs，目的希望實現更加優化的資源分配和管理，特別是應用在超低延遲通訊(URLLC)、增強型移動寬頻通訊(eMBB)和大規模機器類通訊(MMTC)等關鍵5G應用場景下。 在本論文的研究中，有注意BWP切換過程中的延遲問題，並考慮了Inactivity Timer對下行鏈路資源管理策略的影響，提出了兩種不同的BWP切換演算法：一是基於閒置資源的Idle演算法，另一個是基於資源釋放指標的RRI演算法。這兩種演算法的結合BWP Segmentation，使得系統在5G下行鏈路中能夠更靈活地應對各種通訊需求，同時有效管理BWP切換所帶來的潛在延遲，並保持高效的資源利用率和低延遲性能。 本論文進行了一系列實驗和模擬，評估BWP Segmentation方法在不同到達率(Arrival Rate)情境下的表現。實驗結果展示了通過Idle和RRI演算法在5G下行鏈路中實現的延遲減少，也顯示了相對於傳統方法在功耗上的較好優勢。 綜合來看，本論文的研究成果對於解決5G下行鏈路中的資源分配和管理問題，以及滿足不同5G應用場景的需求，具有重要的實際意義。這些成果為建立更高效、可靠和多樣化的5G應用場景提供了幫助，並在節能和減少延遲方面提供了研究方向。 ;This paper investigates an aspect of 5G wireless communication technology: efficient resource allocation and management for the downlink channel. The primary objective is to handle multiple Bandwidth Parts (BWP) effectively to meet the diverse application requirements within the 5G communication system′s downlink channel. In the 5G downlink, BWPs are typically defined by varying Numerologies (denoted by μ), including various subcarrier spacings and time domain configurations. This study introduces a BWP Segmentation approach that aims for optimized resource allocation by combining BWPs based on different Numerology values, particularly for critical 5G applications like Ultra-Reliable Low-Latency Communications (URLLC), Enhanced Mobile Broadband (eMBB), and Massive Machine Type Communications (MMTC). The research pays special attention to the latency involved in BWP switching and examines the impact of the Inactivity Timer on downlink resource management strategies. Two BWP switching algorithms are proposed: one based on idle resources and another based on resource release indicators (RRI). The combination of these algorithms allows for flexible response to communication needs in the 5G downlink and efficient management of the potential latency introduced by BWP switching while maintaining high resource utilization and low latency. Experiments and simulations conducted assess the performance of the BWP Segmentation method under different arrival rate scenarios. Results demonstrate latency reduction achieved through Idle and RRI algorithms and significant power-saving advantages over traditional methods.
Overall, the outcomes of this research are of practical importance for addressing resource allocation and management in the 5G downlink and for meeting the demands of various 5G applications. These findings lay a solid foundation for establishing efficient, reliable, and diversified 5G application scenarios, offering research directions for energy savings and latency reduction.
