A complex random vector is said to be improper if its pseudo-autocorrelation matrix is non-zero. Recently, it has been shown that the performance of linear receivers can be significantly improved if the complex conjugate of the received signal is also utilized in the detection procedure when the received signal is improper and this strategy is called widely linear processing. The received signal becomes improper if transmitter employs real-valued modulation such as binary phase-shift keying (BPSK) or amplitude-shift keying (ASK), or employs space-time block coding (STBC).
In the uplink of MC-CDMA systems, the receiver at the base station typically has the information on the spreading sequences of all users within the cell, but not those of the users in other cells. It is natural to expect that the performance of the receiver can be improved if the information on the spreading sequences of in-cell users is properly utilized.
In this dissertation, we propose three blind adaptive minimum output energy (MOE) receivers for uplink MC-CDMA systems with improper signaling in multi-cell environments. Among them, one is for the real-valued modulation and the others are for the Alamouti`s STBC.
First, a blind adaptive MOE receiver has been proposed for uplink single-input single-output (SISO) MC-CDMA systems employing real-valued modulation in multi-cell environment. In this case, since the imaginary part of residual interference at the filter output does not affect the detection of the desired user`s data, the proposed receiver has been designed to perfectly eliminate the real part of residual in-cell interference using the information on the spreading sequences of all intra-cell users and to reduce the real part of residual inter-cell interference by minimizing the constrained output energy.
Second, a blind adaptive MOE receiver has been proposed for uplink MC-CDMA systems in multi-cell environments in case that all intra-cell users employ STBC and it known at t...