在塊狀時空碼補零系統下基於子空間方法使用少數接收區塊之新盲通道估測演算法

Abstract

在這篇論文中，我們研究在多輸入多輸出補零系統下傳送端使用著Alamouti 塊狀時空編碼的前提下，基於子空間方法之盲通道估測演算法。我們所提出的演算法可以被使用在有著廣義預編碼器之塊狀時空補零系統下，像是單載波補零以及補零正交分頻多工。我們所提出的演算法結合使用了前後平均技術以及循環排列的概念，因此可以在少數接收資料的前提下順利運作。另外，我們使用另一個稱為循環重複 參數。只要適當的選擇這個參數，演算法效能將可以被輕微的改善。且再補零正交分頻多工不使用前後平均技術的前提下，選擇大於零之循環重複參數將可以大幅的提升演算法順利運作的機率。再和前人的方法比較中，可以發現我們有著下列較前人方法優異的地方1) 我們有著更好的通道估測效能; 2) 我們所需的接收區塊較少。且這些優點都將在模擬結果中被證實。

In this thesis, we studied subspace-based blind channel estimation in multiple-input-multiple-output (MIMO) zero-padded (ZP) systems whose transmitter is employed with a special case of space-time-block-codes (STBC), namely, the Alamouti codes. The proposed algorithm can be applied in such STBC-ZP systems with generalized precoders which are equivalent to, for example, zero-padded single carrier (SC-ZP) systems and zero-padded orthogonal frequency division multiplexing (ZP-OFDM) systems. The proposed method is capable of working with a small amount of received data by using the circulant permutation concept and forward and backward (FB) averaging techniques jointly. In the proposed method, we also introduce a new parameter called circulant repetition index. The performance of the proposed method can be slightly improved by choosing a circulant repetition index properly. The probability for the proposed method to work can be further increased by choosing circulant repetition index large than zero in ZP-OFDM systems without using FB technique. Compared with several algorithms in the recent literature, the proposed method possesses advantages in 1) better performance; 2) fewer required received blocks. Simulation results verified these advantages.