請用此 Handle URI 來引用此文件:
http://tdr.lib.ntu.edu.tw/jspui/handle/123456789/83760| 標題: | 混合式波束成形空間預編碼配合最佳化部分連接應用於大規模多輸入多輸出系統 Hybrid Beamforming-Based Spatial Precoding under Optimally Partial Connection for Massive MIMO Systems |
| 作者: | 張景銘 Jing-Ming Chang |
| 指導教授: | 李枝宏 Ju-Hong Lee |
| 關鍵字: | 大規模多輸入多輸出系統,混合式波束成型,空間預編碼,部分連接架構,灰狼最佳化,離散型灰狼最佳化, Massive MIMO,Hybrid beamforming,Spatial precoding,Partially connected,Gary Wolf Optimization,Gary Wolf Optimization, |
| 出版年 : | 2022 |
| 學位: | 碩士 |
| 摘要: | 延續本實驗室過去提出波束成形選擇空間預編碼(Beamforming-selection spatial precoding, BSSP),為了降低毫米波多輸入多輸出系統中的硬體成本,使用BSSP方法結合混合式波束成型(Hybrid Beamforming)預編碼,並使用灰狼最佳化(Grey Wolf Optimization, GWO)演算法調整預編碼的係數。 本篇論文主要改善部分連接(Partially connected)架構在平均位元錯誤率的表現,提出各組通道天線選擇最佳化,以適應在不同通道情況下,傳送端可以選擇最佳的天線選擇方式進行資料傳輸,而部分連接架構又可分為重疊子陣列(Overlapped Subarray, OSA)與不重疊子陣列(Non-overlapped Subarray, NOSA)兩種架構,分別使用二位元灰狼最佳化(Binary Grey Wolf Optimization, BGWO)與離散型灰狼最佳化(Discrete Grey Wolf Optimization, DGWO)兩種離散形最佳化演算法來最佳化天線選擇矩陣。為了使錯誤率更進一步降低,本篇論文將天線選擇最佳化結合多段式最佳化,第一階段最佳化預編碼與天線選擇矩陣,第二階段最佳化接收器,第三階段再最佳化預編碼與天線選擇矩陣,達成傳送端與接收端中,預編碼、天線選擇、接收器都為最佳情形,不但比全連接架構連接更少的天線數目,又可以達到較好的平均位元錯誤率。 Continue to the Beamforming-selection spatial precoding(BBSP) proposed in our laboratory in the past, to reduce the hardware cost in the millimeter-wave multiple-input multiple-output system, the BSSP method combined with the hybrid beamforming precoding is used, and use the Grey Wolf Optimization (GWO) algorithm to adjust the precoding coefficients. This paper mainly improves the performance of the average bit error rate of the partially connected architecture and proposes the optimization of antenna selection for each channel, so that the transmitter can choose the best antenna selection method for data processing under different channel conditions, and the partially connected architecture can be divided into Overlapping Subarray (OSA) and non-overlapped Subarray (NOSA) architectures, respectively using Binary Grey Wolf Optimization (BGWO) and Discrete Grey Wolf Optimization (DGWO) two discrete optimization algorithms to optimize the antenna selection matrix. To further reduce the error rate, this paper combines antenna selection optimization and multi-stage optimization, the first stage optimizes the precoder and antenna selection matrix, the second stage optimizes the receiver and the third stage optimizes the precoder and antenna selection matrix again to achieve precoder, antenna selection, and receiver are all optimal, not only connecting fewer antennas than the fully connected architecture but also achieving a better average bit error rate. |
| URI: | http://tdr.lib.ntu.edu.tw/jspui/handle/123456789/83760 |
| DOI: | 10.6342/NTU202201476 |
| 全文授權: | 未授權 |
| 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 | |
|---|---|---|---|
| ntu-110-2.pdf 未授權公開取用 | 10.31 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。
