請用此 Handle URI 來引用此文件:
http://tdr.lib.ntu.edu.tw/jspui/handle/123456789/54172
標題: | 用於毫米波多輸入多輸出系統之時變通道估計之新方法 New Method for Time-Varying Channel Estimation for Millimeter Wave MIMO Systems |
作者: | Yu-Jhao Yang 楊鈺昭 |
指導教授: | 蘇柏青(Borching Su) |
關鍵字: | 第五代行動通訊系統,毫米波,多輸入多輸出,混合式波束成形,通道狀態資訊,高速移動,時變通道估計,都卜勒偏移, 5G,mmWave,MIMO,hybrid beamforming,CSI,high mobility,time-varying channel estimation,Doppler shift, |
出版年 : | 2020 |
學位: | 碩士 |
摘要: | 毫米波頻率因其高數據速率和豐富的頻譜資源,被視為第五代無線通訊的候選者。多輸入多輸出是五代無線通訊蜂巢式網路的一項可靠的技術,而基於混合波束成形結構的多輸入多輸出架構用於平衡硬體成本和系統性能。在多輸入多輸出系統中設計預編碼器與組合器需要通道狀態資訊,利用毫米波通道的稀疏特性,將通道估測問題轉換成採用基於壓縮感測的演算法來解決問題。為了實現高速移動性環境,必須考慮來自傳輸端和接收端之間相對移動造成的都卜勒偏移,所以必須考慮毫米波多輸入多輸出系統的時變通道估計。 本論文提出了兩種估計都卜勒偏移的方法,方法一利用矩陣形式的通道模型,可以簡單提取都卜勒偏移和通道增益。方法二利用多解析度的概念估測都卜勒偏移。模擬結果顯示方法一在訊噪比高的時候表現良好,同時擁有最低的計算複雜度,方法二擁有最好的表現。 Millimeter wave (mmWave) frequencies have been envisioned as a promising candidate for fifth-generation (5G) wireless communications due to its high data rate and rich spectrum resources. Multiple-input multiple-output (MIMO) is a promising technique for 5G cellular networks, a MIMO architecture based on hybrid beamforming structure is used to strike a tradeoff between hardware cost and system performance. Channel state information (CSI) is needed to design precoder and combiner in MIMO systems, by leveraging the sparse nature of mmWave channel, channel estimation is formulated as a sparse problem, compressed sensing (CS) based algorithm is used to solve the problem. For achieving the high mobility environment, it is necessary to consider the Doppler shift which originating from relative movements between TX and RX, so that a time-varying channel estimation for mmWave MIMO systems should be considered. In this thesis, two new Doppler shift estimation are proposed. For proposed method 1, by exploiting the channel model with matrix form, we can extract Doppler shift and complex channel gains in a simple way. For proposed method 2, by utilizing multi-resolution searching of Doppler shift, and then estimate Doppler shift. Simulation results show that proposed method 1 has a good performance with high $E_{b}/N_{0}$, and has lower computation. Proposed method 2 has the best performance. |
URI: | http://tdr.lib.ntu.edu.tw/jspui/handle/123456789/54172 |
DOI: | 10.6342/NTU202002316 |
全文授權: | 有償授權 |
顯示於系所單位: | 電信工程學研究所 |
文件中的檔案:
檔案 | 大小 | 格式 | |
---|---|---|---|
U0001-0308202021064200.pdf 目前未授權公開取用 | 1.75 MB | Adobe PDF |
系統中的文件,除了特別指名其著作權條款之外,均受到著作權保護,並且保留所有的權利。