OFDM transmission has been proposed in the literature for relay nodes to combat the time delays from the relay nodes, where the paths from relay nodes to destination node are treated as multipaths and space-time (or frequency) coding is used to achieve the spatial (or multipath) diversity. With this approach, when the cyclic prefix (CP) length is less than the time delay length, inter-block interference occurs. In this paper, we consider Alamouti coded OFDM systems in cooperative communications where the CP length may be less than the time delay length. By taking the advantage of the Alamouti code structure, we propose a time domain interference cancellation algorithm. Simulation results show that the performance is significantly improved with our newly proposed time domain interference cancellation algorithm.

An efficient way to reduce the peak-to-average power ratio (PAPR) in OFDM systems is clipping. After the clipping in an MIMO-OFDM system, the additive noise may not be white. In this paper, we develop fast (single-symbol) ML decoding algorithms for orthogonal space-time block codes (OSTBC) and (linearly transformed) quasi orthogonal space- time block codes (QOSTBC) in clipped MIMO-OFDM systems by using a clipping noise model with Gaussian approximation. By using the statistics of the clipping distortions, our newly developed fast ML decoding algorithms improve the performance for clipped MIMO-OFDM systems with OSTBC and QOSTBC while the decoding complexities are not increased. Simulation results are presented to illustrate the improvement.

Space-time/frequency coding (SFC) can achieve the spatial and multipath diversities for a MIMO-OFDM system by coding across subcarriers, multiple antennas, and/or multiple OFDM sysmbols, where an interesting method to achieve the multipath diversity is repeating across subcarriers proposed by Su et al. While most of the existing space-time/frequency codes do not have the fast ML decoding, a family of space-time-frequency codes with single-symbol ML decoding have been proposed lately by Zhang et al to achieve both full spatial and multipath diversities by using orthogonal space-time block codes (OSTBC) across multiple antennas and OFDM symbols and in the meantime repeating across the subcarriers. In this paper, we first generalize the above OSTBC to linearly transformed quasi OSTBC (QOSTBC) in a straightforward way. The main goal of this paper is to modify the repeating process and adjust their phases so that the peak-to-average power ratio (PAPR) of the OFDM system is reduced. In particular, we propose to use Chu sequences and show that the discrete PAPR can be reduced by Gamma times where Gamma is the times of the repeating across subcarriers for any SFC from the repeating.

In a clipped MIMO-OFDM system, the overall additive noise, including the clipping distortion, may not be white. In this paper, we develop fast ML decoding algorithms for orthogonal space-time-frequency block codes (OSTFBC) and quasi orthogonal space-time-frequency block codes (QOSTFBC) in clipped MIMO-OFDM systems by using a clipping noise model with Gaussian approximation. By using the statistics of the clipping distortions, our newly proposed fast ML decoding algorithms improve the performance for clipped MIMO-OFDM systems with OSTFBC and QOSTFBC without increasing the decoding complexity. Simulation results are presented to illustrate the improvement.

Space-time/frequency coding (SFC) can achieve the spatial and multipath diversities for an MIMO-OFDM system by coding across subcarriers, multiple antennas, and/or multiple OFDM symbols, where an interesting method to achieve the multipath diversity is repeating across subcarriers proposed by Su et al. Lately, a family of space-time-frequency codes with single-symbol ML decoding have been proposed by Zhang et al to achieve both full spatial and multipath diversities by using orthogonal space-time block codes (OSTBC) across multiple antennas and OFDM symbols and in the meantime repeating across the subcarriers. The main goal of this paper is to modify the repeating process and adjust their phases by a Chu sequence so that the peak-to-average power ratio (PAPR) of the OFDM system is reduced for any SFC from the repeating and in the meantime, the full spatial and multipath diversities are maintained. In this paper, we also generalize the space-time- frequency coding by Zhang et al from using OSTBC to using linearly transformed quasi-orthogonal space-time block codes (QOSTBC) that possesses higher rate than the OSTBC for more than two transmit antennas, fast ML decoding and full spatial and multipath diversities.

An efficient way to reduce the peak-to-average power ratio (PAPR) in OFDM systems is clipping. After the clipping in an MIMO-OFDM system, the additive noise may not be white. In this paper, we develop fast ML decoding algorithms for orthogonal space-time block codes (OSTBC) and quasi orthogonal space-time block codes (QOSTBC) in clipped MIMO-OFDM systems by using a clipping noise model with Gaussian approximation. By using the statistics of the clipping distortions, our newly proposed fast ML decoding algorithms improve the performance for clipped MIMO-OFDM systems with OSTBC and QOSTBC without increasing the decoding complexity. Simulation results are presented to illustrate the improvement.