City University of Hong Kong

CityU Institutional Repository >
3_CityU Electronic Theses and Dissertations >
ETD - Dept. of Electronic Engineering  >
EE - Master of Philosophy  >

Please use this identifier to cite or link to this item:

Title: Channel estimation and equalization for OFDM over doubly selective channels
Other Titles: Zheng jiao pin fen fu yong xi tong zai shuang xuan xin dao zhong de xin dao gu ji he jun heng
Authors: Yu, Yao (于瑤)
Department: Dept. of Electronic Engineering
Degree: Master of Philosophy
Issue Date: 2007
Publisher: City University of Hong Kong
Subjects: Orthogonal frequency division multiplexing
Wireless communication systems
Notes: CityU Call Number: TK5103.484.Y88 2007
Includes bibliographical references (leaves 111-116)
Thesis (M.Phil.)--City University of Hong Kong, 2007
xi, 117 leaves : ill. ; 30 cm.
Type: Thesis
Abstract: Wireless communication systems at high-data rates and high terminal speeds are currently highly demanded. High data rate and high terminal speed induce severe intersymbol interference (ISI) in band-limited channels and inter-carrier interference (ICI) in multi-carrier systems. Wireless channel with multipath for high terminal mobility is generally a frequency-selective and time-selective channel, called as doubly selective channel (DSC). Orthogonal Frequency Division Multiplexing (OFDM) is a multi-carrier system that can efficiently remove ISI by using cyclic prefix and converting a frequency selective channel into several frequency non-selective sub-channels. However, the ICI caused by the time selectivity of doubly selective channels (DSCs) destroys the orthogonality between OFDM subcarriers and thus degrades the system performance. Therefore, channel estimation and equalization are of vital importance for OFDM systems over DSCs. In this thesis, we propose an iterative frequency-domain cancellation equalizer (FCE) to suppress the ICI effects. The channel frequency response matrix (CFRM) is divided into two parts: target CFRM, which is composed of the main diagonals of the CFRM, and interference CFRM, which contains the remaining elements of the CFRM. In the iterative equalization process, the interference signals generated from the interference CFRM are subtracted from the received signal, resulting in a partial-interference-cancelled signal. Using the target CFRM with the partial-interference-cancelled signal obtains a small scale linear minimum mean square error (LMMSE) equalizer and a signal estimate for individual subcarrier iteratively. Simulation results show that this approach outperforms the standard LMMSE with much lower complexity than LMMSE and achieve the better performance than the iterative frequency-domain windowed equalizer (FWE) while having the complexity comparable to FWE. Moreover, we modify and apply an iterative time-domain equalizer (TDE) designed originally for single-carrier systems to OFDM systems. This iterative equalizer equalizes DSCs in the time domain, which makes use of the exactly banded structure of channel impulse response matrix (CIRM) and bypasses the construction of CFRM using basis parameters obtained in the time domain. Although the performance comparison between FCE, FWE and TDE depends on different factors, FCE can generally yield the better BER performance than TDE if FCE selects a proper bandwidth; nevertheless, the computational complexities of these methods are comparable. An iterative channel estimation and equalization for OFDM with pilots over DSCs is proposed. In the OFDM system, several short pilot sequences are inserted into OFDM symbols. The received signals at the pilot subcarriers are employed to estimate the initial channel coefficients. With these initial channel coefficients, we obtain the initial estimates of OFDM symbols through iterative channel equalization. Next the detected symbols are all applied to perform channel estimation. It is shown that this approach can yield good channel estimation and BER performance. The simulation results of joint channel estimation and three different equalization methods: FCE, FWE and TDE demonstrate that the normalized MSE of the proposed channel estimation can approach to that of channel estimation with the whole block for training. Moreover, the simulation results show that the proposed joint channel estimation and equalization using FCE provides smaller BER than that of using FWE and TDE. Finally, the three iterative equalization methods, FCE, FWE and TDE, are compared and analyzed theoretically. Analysis shows that FCE can provide better signalto- interference-plus-noise ratio (SINR) than FWE, which explains why FCE yields the better BER performance than FWE. On the other hand, the comparison between FCE and TDE is complicated. Although MSE of TDE is much smaller than that of FCE, TDE cannot correct the large burst errors by the iterative equalization. Therefore, the BER performance of FCE is better than TDE with a proper band size. In conclusion, we propose an iterative receiver joint channel estimation and equalization both in the frequency domain and time domain for OFDM systems over DSCs. The new receiver can efficiently suppress the ICI effects and achieve better performance than some existing methods with comparable computational complexity.
Online Catalog Link:
Appears in Collections:EE - Master of Philosophy

Files in This Item:

File Description SizeFormat
fulltext.html159 BHTMLView/Open
abstract.html159 BHTMLView/Open

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


Valid XHTML 1.0!
DSpace Software © 2013 CityU Library - Send feedback to Library Systems
Privacy Policy · Copyright · Disclaimer