| Summary: | 碩士 === 輔仁大學 === 電子工程學系 === 94 === Recently, the space-time coded multi-carrier code-division multiple-access (ST-coded MC-CDMA) has been widely investigated for the great demand on capacities and transmission rates in the future wireless communication systems. In this chapter, we study the blind channel estimation and multiuser detection for quasi-synchronous multi-carrier code-division multiple-access (MC-CDMA) multiple-input multiple-output (MIMO) system with 4-Tx space-time block codes (STBC). The subspace-based channel estimation has been successfully for different CDMA systems. Here we present a subspace-based channel estimator which utilizes the space-time coding property to improve the performance of channel estimator. By either singular value decomposition (SVD) on the data matrix or eigenvalue decomposition (EVD) on the correlation matrix, the noise subspace can be obtained. Then the channel impulse response is estimated by the quadratic optimization problem. We investigate the performance analysis using the first order perturbation theory on the EVD of correlation matrix is derived. After obtaining the channel impulse response, the multiuser detection is developed afterwards. We adopt the eigenspace technique and the forward-backward techniques to reduce the performance degradation due to finite samples and/or vector errors. In order to get faster convergence rate, the weight vectors of these four filters are shown to be related each other, and then only about quarter of the computational complexity of eigenspace filter can be reduced.
|
|---|
