應用於LTE收發器之多輸入多輸出正交分頻多工系統的可適性通道估測設計

Abstract

本論文研製了一個可應用於3GPP Long-Term Evolution(LTE) 無線網路標準的可適性通道估計演算法，此可適性通道估測演算法可對抗時變之頻率選擇性快速衰減通道。LTE無線網路標準為設計時的參考平台，2對2的多重輸入多重輸出正交分頻多工系統被納入考慮。 本論文提出的通道估測演算法包含兩個步驟。第一個步驟利用PDCCH初步估測通道後，再經由本論文所提出的可適性通道估計演算法做處理，以估計出包含雜訊的時變通道，再由頻域上的一階低通濾波器濾去部分的雜訊，以降低演算法對訊噪比的需求。 根據模擬的結果，此可適性等化器在2對2的多輸入多輸出系統下，可在傑克斯模型120km/hr、訊噪比33dB的條件下可達到與One-shot通道估測演算法相同的效能，且與One-shot通道估測演算法比較，乘法數比One-shot通道估測演算法減少88%，且記憶體的容量為One-shot通道估測演算法的1/14。

An adaptive channel estimation algorithm applied to 3GPP Long-Term Evolution (LTE) protocol under time-variant frequency-selective fast fading channels is proposed in this thesis. LTE wireless internet protocol is the design reference and 2x2 multiple inputs multiple outputs (MIMO) Orthogonal Frequency Division Multiplexing (OFDM) are considered. There are two steps in the presented method. Firstly, the system estimates the channels by PDCCH. Secondly, the proposed adaptive channel estimation is applied to SFBC-decoded symbols in order to estimate the noisy time-variant channels, and a frequency domain first order low pass filter is applied to each estimated noisy time-variant channel in order to decrease the Signal-to-Noise Ratio (SNR) requirements that adaptive equalization requires. As simulation result, the proposed adaptive channel estimation under 120km/hr Jakes’ model at 33dB SNR for 2 2 MIMO-OFDM systems have the same performance as One-shot channel estimation. Compare with One-shot channel estimation, the proposed algorithm has reduced 88% multiplier in channel estimate, and the memory size of proposed algorithm is 1/14 of One-shot channel estimations’.