In this paper, partial data-dependent superimposed training based channel estimation for OFDM systems over doubly selective channels (DSCs) is addressed. Due to the presence of unknown data as interference, we first derive a minimum mean square error (MMSE) channel estimator by treating the effect of unknown data as noise. To further improve the performance, a novel iterative algorithm which jointly estimates channel and suppresses interference from data is proposed via variational inference approach. Simulation results show that the proposed algorithm converges after a few iterations. Furthermore, after convergence, the performance of the proposed channel estimator is very close to that with full training at high SNRs.

Semi-blind joint CFO, channel estimation and data detection for OFDM systems over doubly selective channels (DSCs) is investigated in this work. A joint iterative algorithm is developed based on the maximum a posteriori expectation-maximization (MAP-EM) algorithm. In addition, a novel algorithm is also proposed to obtain the initial estimates of CFO and channels. Simulation results show that the performance of the proposed CFO and channel estimators approaches to that of the estimators with full training at high SNRs. Moreover, after convergence, the performance of data detection is close to the ideal case with perfect CFO and channel state information.

Data detection for OFDM systems over unknown doubly selective channels (DSCs) and carrier frequency offset (CFO) is investigated. A semiblind iterative detection algorithm is developed based on the expectation-maximization (EM) algorithm. It iteratively estimates the CFO, channel and recovers the unknown data using only limited number of pilot subcarriers in one OFDM symbol. In addition, efficient initial CFO and channel estimates are also derived based on approximated maximum likelihood (ML) and minimum mean square error (MMSE) criteria respectively. Simulation results show that the proposed data detection algorithm converges in a few iterations and moreover, its performance is close to the ideal case with perfect CFO and channel state information.

In this paper, a joint channel estimation and data detection algorithm is proposed for OFDM systems under doubly selective channels (DSCs). After representing the DSC using Karhunen-Loe¿ve basis expansion model (K-L BEM), the proposed algorithm is developed based on the expectation-maximization (EM) algorithm. Basically, it is an iterative algorithm including two steps at each iteration. In the first step, the unknown coefficients in K-L BEM are first integrated out to obtain a function which only depends on data, and meanwhile, a maximum a posteriori (MAP) channel estimator is obtained. In the second step, data are directly detected by a novel approach based on the function obtained in the first step. Moreover, a Bayesian Cramer-Rao Lower Bound (BCRB) which is valid for any channel estimator is also derived to evaluate the performance of the proposed channel estimator. The effectiveness of the proposed algorithm is finally corroborated by simulation results.