| Summary: | 碩士 === 國防大學中正理工學院 === 電子工程研究所 === 91 === This thesis deals with multiuser detection under spreading code mismatch and carrier frequency offset for multicarrier code division multiple access system. In spreading code mismatch case, to cope with the problems of both spreading code mismatch and computing cost, we investigate two blind approaches to multiple access interference (MAI) suppression: the robust eigenspace-based technique and the linearly constrained minor component analysis. These methods are robust to the code mismatch, not only because they can correct the spreading code mismatch, but also can reduce computing cost and have fast convergence. Many blind approaches have been proposed to estimate the carrier frequency offset by applying the characteristic of the orthogonality between virtual carriers and information carrying carriers. But these methods are sensitive to noise and suffer to more computational complexity for high-resolution estimate. Two estimators for correcting carrier frequency offset with improved robustness to noise are also presented in this thesis. One is Gram-Schmidt orthogonalization estimator and the other one is projection estimator. Several simulation results show that these proposed estimators not only can estimate and correct frequency offset, but also can reduce computing cost.
|
|---|
