Blind Channel Estimation For Orthogonal Space Time Block Codes In Miso Systems Microform

This thesis presents a closed-form, blind channel estimation scheme for Alamouti's and other orthogonal STBC. Unlike other blind channel estimation algorithms used in such systems, the scheme is able to estimate the channels, to within a phase constant, in multiple-input single-output systems, i.e. systems that employ only one receive antenna. The channel matrix is estimated from the eigenvalue decomposition of the fourth order cumulant matrix of the output signal. The performance of the scheme depends partly upon the accuracy of the estimated cumulants, and thus a scheme to improve the cumulant matrix estimate is suggested. Using these improved cumulants, the algorithm performs very well in slowly changing channel conditions. A single pilot-tuple is required to correctly assign the estimated to the actual channels and to resolve the phase ambiguity common to all blind estimators. The main disadvantage of the scheme is its higher complexity related to the estimation of higher order cumulants; this complexity can be reduced by exploiting the symmetry inherent in the cumulant matrix. Furthermore, it is shown that to achieve good performance in terms of bit error rate, 1000 sample points are sufficient to estimate the cumulant matrix.