LTE(Long Term Evolution)系統是由3GPP(3rd Generation Partnership Project)所提出的第四代行動通訊系統。LTE下行鏈路採用了OFDMA(Orthogonal Frequency Division Multiple Access)系統，它是根據OFDM(Orthogonal Frequency Division Multiplexing)系統演進而來，因此與OFDM面臨了相同的挑戰。其中最具挑戰性的問題之一是在隨著時間變化的通道環境中，而引起子載波間正交性被破壞及載波間干擾(Inter-Carrier- Interference，ICI)。此外，因通道會隨著時間變化的特性，而需要更多的參考訊號密集的分佈在時域的子載波上，以利於通道估測，卻使得頻譜效率大幅降低。 在本篇論文中，我們將提出LTE下行鏈路系統在時間變化的環境中，改善其通道估測的準確度，傳統的通道估測方式使用多項式擴展模型(P-BEM)可以直接實現在LTE下行鏈路系統中的參考訊號擺放方式，然而這些方法在時變通道的環境下，因頻域與時域上通道的變化使其性能大幅下降。在此篇論文中，我們提出了一種改進通道估測的方式，結合了資料直接決策(Data-Decision Directed)與多項式擴展模型(P-BEM)的機制，透過模擬來證明通道估測性能之改善。 LTE (Long Term Evolution) system is the fourth generation mobile communication system proposed by the 3GPP (3rd Generation Partnership Project). In the down link, LTE system adopts Orthogonal Frequency Division Multiple Access (OFDMA) technique which is based on Orthogonal Frequency Division Multiplexing (OFDM) and therefore faces the same challenging problems. One of the most challenging problem is the channel estimation in the time-varying environments where the orthogonality between the subcarriers is destroyed and inter-carrier interference (ICI) is caused. Moreover the time-varying channel characteristic implies a denser pilot subcarriers distribution in time-domain when the simple pilot-aided channel estimation is applied, which incurs a decrease in spectral efficiency. In this thesis, we tackle the channel estimation problem in time-varying environments specifically for LTE downlink system. Traditionally the channel estimation can be directly achieved with the polynomial basis expansion modeling methods (P-BEM) based on the pilot pattern specified in LTE downlink system. However, the performance of these methods degrades in time-varying environments due to the channel variation in both frequency domain and time domain. In this thesis, we propose an improvement method which incorporates data-decision directed mechanism with the P-BEM. The performance improvement is demonstrated through simulation with a cost in computation complexity.
