用於正交分頻多工系統中之低複雜度的線性最小均方差通道估測法

Abstract

在正交分頻多工系統中，使用線性最小均方差法(LMMSE)在估測通道上有很好的效果乃是眾所皆知的。然而，使用此法必須知道通道的統計特性且此法的計算複雜度是非常高的。在本碩士論文中，我們提出一低複雜度的線性最小均方差法來解決此問題。首先，我們先利用最小平方法(LS)來粗估頻域上的通道響應。然後，將此粗估的結果轉換到時域上，再進一步去偵測非零項的通道位置。最後，這些被偵測到非零項的通道再經過濾波器做進一步的處理，而最佳的濾波器係數可藉由最小均方差法的準則得到。這個通道估測器並不需要知道通道的統計特性且計算複雜度也較低。由電腦模擬的結果顯示，所提方法的效能幾乎跟有真正通道響應的效能是很接近的。

It is well known that the linear minimum mean square (LMMSE) channel estimate for orthogonal frequency division multiplexing (OFDM) systems has good performance. However, it requires the channel statistics and the computational complexity is high. In this thesis, we propose a low-complexity LMMSE channel estimator to solve the problems. First, the least-squares method is used to obtain a channel estimate in the frequency domain. Then, the channel estimate is transformed to the time domain and non-zero channel taps are detected. Finally, the least-squares solutions for those non-zero taps are filtered and the optimal filter weights are obtained using a minimum mean square (MMSE) criterion. This channel estimator does not require channel information and the computational complexity is low. Simulations results show that the performance of the proposed channel estimator is almost identical to the case where the channel response is exactly known.