Please use this identifier to cite or link to this item: http://hdl.handle.net/11144/3704
Title: Impact of Channel Estimation Errors on SC-FDE Systems
Authors: Dinis, Rui
Silva, João
Souto, Nuno
Keywords: Channel estimation
Frequency-domain equalization
Iterative equalization
OFDM
SC-FDE
Issue Date: 2014
Abstract: Single carrier transmissions with frequency domain equalization (SC-FDE) have gained widespread use in emergent broadband wireless systems becoming an attractive alternative to popular Orthogonal Frequency Division Multiplexing (OFDM) schemes, particularly at the uplink. Since coherent receivers are usually employed with SC-FDE, accurate channel estimates are required so as to avoid substantial performance degradation. Several channel estimation strategies have been proposed for SC-FDE, but a thoroughly evaluation of the degradation caused by channel estimation errors and a comparison against OFDM is still lacking. In this paper we study the impact of imperfect channel knowledge on SC transmission with focus on the linear frequency domain equalizer (FDE) and on the Iterative Block Decision Feedback Equalizer (IB-DFE). We propose a modified IB-DFE which incorporates knowledge of the channel estimation error model and show that its performance becomes more robust against the presence of strong error components in the channel estimates. We also evaluate, analytically and through simulations, the degradation caused by imperfect channel estimation in SC-FDE and compare it against OFDM schemes (Orthogonal Frequency Division Multiplexing). It is shown that the channel estimation requirements for SC-FDE are higher than for OFDM unless a channel estimation error aware receiver is employed.
Peer reviewed: yes
URI: http://hdl.handle.net/11144/3704
ISSN: 0090-6778
Appears in Collections:DCT - Artigos/Papers

Files in This Item:
File Description SizeFormat 
impactchannel.pdf6.66 MBAdobe PDFView/Open


FacebookTwitterDeliciousLinkedInDiggGoogle BookmarksMySpace
Formato BibTex MendeleyEndnote Currículo DeGóis 

Items in DSpace are protected by copyright, with all rights reserved, unless otherwise indicated.