
For FullText PDF, please login, if you are a member of IEICE,
or go to Pay Per View on menu list, if you are a nonmember of IEICE.

Joint Signal Detection and Channel Estimation Using Differential Models via EM Algorithm for OFDM Mobile Communications
Kazushi MURAOKA Kazuhiko FUKAWA Hiroshi SUZUKI Satoshi SUYAMA
Publication
IEICE TRANSACTIONS on Communications
Vol.E94B
No.2
pp.533545 Publication Date: 2011/02/01
Online ISSN: 17451345
DOI: 10.1587/transcom.E94.B.533
Print ISSN: 09168516 Type of Manuscript: PAPER Category: Wireless Communication Technologies Keyword: mobile communication, OFDM, EM algorithm, channel estimation, Kalman filter, differential model,
Full Text: PDF>>
Summary:
This paper proposes a new approach for the joint processing of signal detection and channel estimation based on the expectationmaximization (EM) algorithm in orthogonal frequency division multiplexing (OFDM) mobile communications. Conventional schemes based on the EM algorithm estimate a channel impulse response using Kalman filter, and employ the random walk model or the firstorder autoregressive (AR) model to derive the process equation for the filter. Since these models assume that the timevariation of the impulse response is white noise without considering any autocorrelation property, the accuracy of the channel estimation deteriorates under fastfading conditions, resulting in an increased packet error rate (PER). To improve the accuracy of the estimation of fastfading channels, the proposed scheme employs a differential model that allows the correlated timevariation to be considered by introducing the first and higherorder time differentials of the channel impulse response. In addition, this paper derives a forward recursive form of the channel estimation along both the frequency and time axes in order to reduce the computational complexity. Computer simulations of channels under fast multipath fading conditions demonstrate that the proposed method is superior in PER to the conventional schemes that employ the random walk model.

