Doubly Iterative Receiver for Block Transmissions with EM-based Channel Estimation.

The-Hanh Pham,Ying-Chang Liang,A. Nallanathan

DOI: https://doi.org/10.1109/twc.2009.071416

IF: 10.4

2009-01-01

IEEE Transactions on Wireless Communications

Abstract:Cyclic-prefix code division multiple access (CP-CDMA), multicarrier CDMA (MC-CDMA) and single carrier cyclic-prefix (SCCP) transmission are some schemes that could support the increasing demand of future high data rate applications. The linear and nonlinear equalizers used to detect the transmitted signal are always far from the maximum-likelihood (ML) detection bound. The block iterative generalized decision feedback equalizer (BI-GDFE) is an iterative and effective interference cancelation scheme which could provide near-ML performance yet with very low complexity. In order to deploy this scheme, the channel state information (CSI) must be available at the receiver. In practice, this information has to be estimated by using pilot and data symbols. This paper investigates the problem of channel estimation using the expectation maximization (EM) algorithm. The BI-GDFE provides the soft information of the transmitted signals to the EM-based algorithm in the form a combination of hard decision and a coefficient so-called the input-decision correlation (IDC). The resultant receiver becomes a doubly iterative scheme. To evaluate the performance of the proposed estimation algorithm, the Cramer-Rao lower bound (CRLB) is also derived. Computer simulations show that the bit error rate (BER) performance of the proposed receiver for joint channel estimation and signal detection can reach the performance of the BI-GDFE with perfect CSI.