用於混合式波束成形正交分頻多工系統之高維度通道張量估測演算法;Higher-Order Tensor-Based Channel Estimation in Hybrid Beamforming MIMO-OFDM Systems

NCUIR > College of Electrical Engineering & Computer Science > Graduate Institute of Electrical Engineering > Electronic Thesis & Dissertation > Item 987654321/95831

Please use this identifier to cite or link to this item: http://ir.lib.ncu.edu.tw/handle/987654321/95831

Title:	用於混合式波束成形正交分頻多工系統之高維度通道張量估測演算法;Higher-Order Tensor-Based Channel Estimation in Hybrid Beamforming MIMO-OFDM Systems
Authors:	柯翔展;Ke, Hsiang-Chan
Contributors:	電機工程學系
Keywords:	通道估測;正交匹配追蹤;克拉馬－羅下限;混和式波束成形;多輸入多輸出正交分頻多工系統;channel estimation;orthogonal matching pursuit (OMP);Cramer-Rao bound;hybrid beamforming;MIMO-OFDM
Date:	2024-08-14
Issue Date:	2024-10-09 17:19:08 (UTC+8)
Publisher:	國立中央大學
Abstract:	本論文主要對於高維度通道張量估測演算法進行研究，使用的系統模型為單使用者多輸入多輸出(multi-input multi-output, MIMO)正交分頻多工(orthogonal frequency-division multiplexing, OFDM)與採用均勻矩形天線陣列(uniform rectangular array, URA)之混和式波束成形(hybrid beamforming)的架構，並且對於所構成的5維的通道張量進行估測。首先我們對於均勻矩形陣列的天線擺放方式提出了兩種依據角度分析方式不同的波束成形架構，並比較兩種架構下通道估測演算法的表現及複雜度差異。此外我們提出了干擾消除演算法，解決了旋轉張量正交匹配追蹤(Tensor-Orthogonal Matching Pursuit with Rotation, T-OMP-R)演算法中，鄰近路徑間的相互干擾問題，使演算法的估測精準度有著非常大的提升。再來我們提出了不需要繁雜的偽逆運算(pseudo inverse)路徑增益計算方式，加上以階層式離散傅立葉轉換碼簿(hierarchical discrete Fourier transform codebook)取代原先的循序搜尋法，在維持演算法表現的同時大幅度地節省了運算複雜度。為了進一步確保演算法的正確性及表現，我們對於提出的兩種波束成形架構下的演算法均進行克拉馬－羅下限(Cramer-Rao Bound, CRB)以及正規化平均平方誤差(normalized mean square error, NMSE)的分析，分析不同的信號雜訊比(signal-to-noise ratio, SNR)下對於演算法的影響，並與相關通道張量估測演算法進行比較，比較不同的優化方式對於運算複雜度的影響。最後我們對於我們所提出之演算法進行硬體實現上的評估與運算化簡，節省了約41%的實數乘法次數以及約95%的所需時脈。;This thesis studies higher-order tensor-based channel estimation, in single-user hybrid beamforming multi-input multi-output (MIMO) orthogonal frequency-division multiplexing (OFDM) systems with uniform rectangular array (URA) at the transmitter and the receiver. Two hybrid beamforming patterns are proposed for channel estimation. Additionally, we introduce an interference cancellation algorithm to address the issue of mutual interference between adjacent paths in the Tensor-Orthogonal Matching Pursuit with Rotation (T-OMP-R) algorithm, the proposed algorithm significantly improves the accuracy. We further propose a pseudo-inverse free path gain calculation method, which eliminate about 10% computation complexity. Furthermore, by replacing the original sequential searching with a hierarchical discrete Fourier transform codebook, we substantially reduce the arithmetic complexity while maintaining the algorithm performance. To ensure the validity and performance of the proposed algorithm, we conduct analysis using the Cramer-Rao bound (CRB) to examine the estimation performance of the algorithms. Compared to some prior works applied in 5D tensor-based channel estimation, our proposed algorithm achieves similar performance with much reduced complexity. Moreover, the hardware architecture is also designed and evaluated. With computation reduction scheme, 41% multiplication operations and 95% computation cycles can be saved.
Appears in Collections:	[Graduate Institute of Electrical Engineering] Electronic Thesis & Dissertation

Files in This Item:

File	Description	Size	Format
index.html		0Kb	HTML	23	View/Open

社群 sharing

Loading...