Phase noise estimation and mitigation for OFDM systems
Document Type
Article
Publication Date
12-1-2006
Abstract
OFDM suffers from severe performance degradation in the presence of phase noise. In particular, phase noise leads to common phase error (CPE) as well as intercarrier interference (ICI) in the frequency domain. Some approaches in the literature mitigate phase noise by directly evaluating and then compensating for CPE or ICI, while others choose to correct phase noise in the time domain. A new parametric model of OFDM signals is proposed in this paper which shows that, in the presence of phase noise, each received frequency-domain subcarrier signal can be expressed as a sum of all subcarrier signals weighted by a vector parameter. Then, two reduced-complexity techniques are presented to estimate this weighting vector. The first is a maximum likelihood (ML) method whereas the second one is a linear minimum mean square error (LMMSE) technique. Using the obtained estimates, we also propose two approaches, i.e., a decorrelator and an interference canceler, to mitigate phase noise. It is shown that most conventional methods can be readily obtained from our approaches with some approximation or orthogonal transform. Theoretical analysis and numerical results are provided to elaborate the proposed schemes. We show that the performance of both approaches is superior to that of conventional methods. Furthermore, LMMSE gives the best performance, while ML provides a much simpler yet effective way to mitigate phase noise. © 2006 IEEE.
Identifier
33845622953 (Scopus)
Publication Title
IEEE Transactions on Wireless Communications
External Full Text Location
https://doi.org/10.1109/TWC.2006.256984
ISSN
15361276
First Page
3616
Last Page
3625
Issue
12
Volume
5
Grant
CCR-0085846
Fund Ref
National Science Foundation
Recommended Citation
Wu, Songping; Liu, Pan; and Bar-Ness, Yeheskel, "Phase noise estimation and mitigation for OFDM systems" (2006). Faculty Publications. 18693.
https://digitalcommons.njit.edu/fac_pubs/18693
