最大期望演算法用於正交分頻多工之相位雜訊移除與通道估計 ; An EM Algorithm for Phase Noise Removal and Channel Estimation in OFDM Systems

NCUIR > College of Electrical Engineering & Computer Science > Graduate Institute of Communication Engineering > Electronic Thesis & Dissertation > Item 987654321/44519

Please use this identifier to cite or link to this item: http://ir.lib.ncu.edu.tw/handle/987654321/44519

Title:	最大期望演算法用於正交分頻多工之相位雜訊移除與通道估計;An EM Algorithm for Phase Noise Removal and Channel Estimation in OFDM Systems
Authors:	楊善任;Shan-Ren Yang
Contributors:	通訊工程研究所
Keywords:	相位雜訊;最大期望演算法;maximum-expected (EM) algorithm;phase noise
Date:	2010-08-17
Issue Date:	2010-12-09 13:47:06 (UTC+8)
Publisher:	國立中央大學
Abstract:	在正交分頻多工系統中，由相位雜訊所產生的共同相位錯誤與子載波間干擾，會影響載波間正交性並大幅降低系統效能。於許多文獻中已提出一些解決方式，在我們的研究中，利用最大期望演算法估計相位雜訊並抑制雜訊的干擾，以求得更準確的相位雜訊估計值。為了應付在時變通道造成的通道效應，加入決策回授機制，但由於決策回授機制有其錯誤傳播效應，因此我們結合最大期望演算法的通道估測方式彌補決策回授機制在高雜訊下準確性的不足。而從訊號干擾比與位元錯誤率的模擬分析中，亦可以看出我們演算法效能較為優異，而在時變通道中，我們所提出的方式可以有效的偵測通道變化，降低位元錯誤率。The common phase error (CPE) and the intercarrier interference (ICI) due to the phase noise arisen in an orthogonal frequency division multiplexing (OFDM) system can significantly reduce system performance. In contrast to previous proposals in the literature, we study the maximum-expected (EM) algorithm to obtain more accurate estimate of the phase noise and to suppress the resulting ICI. To consider the application in a time-varying channel environment, the channel estimation method is implemented by the combination of the EM algorithm and a decision feedback method to take advantage of both merits in order to prevent from the influence of decision errors at low SNR. From performance analysis and simulations of SINR and BER for the proposed algorithm, the simulated SINR results are close to the analyzed ones and the BER performance has better result than the previous work.
Appears in Collections:	[Graduate Institute of Communication Engineering] Electronic Thesis & Dissertation

Files in This Item:

File	Description	Size	Format
index.html		0Kb	HTML	865	View/Open

社群 sharing

Loading...