Study of Transmit Beamforming Methods for MIMO Systems

[[abstract]]在本論文中，我們在多輸入多輸出(multiple-input multiple-output, MIMO)系統下於傳送端使用傳送波束形成(Transmit Beamforming)技術，在相同環境下探討四種權重決定演算法：奇異值分解(Singular Value Decomposition，SVD)、最大範數結合(Maximum Norm Combining，MNC)、最大比率傳送(Maximum Ration Transmit, MRT)以及改良最大比例傳送(Improved Maximum Ration Transmit, IMRT)的系統效能，模擬並探討四者的在不同天線數下的系統效能與複雜度。 我們透過在雷利衰減通道(Rayleigh fading channel)下做系統模擬，並比較四種權重決定法在相同訊雜比(Signal to Noise Ratio, SNR)下的位元錯誤率(Bit Error Rate，BER)。在傳送端天線數量低的情況下，四種演算法的系統效能相當接近，但是MRT需要考慮接收端天線數，雖然複雜度低，隨著接收天線數上升效能會下降，而在天線數量多的情況下，在系統效能及複雜度間的取捨中選擇MNC會比SVD好。[[abstract]]We study the transmit Beamforming technique for MIMO system in this thesis, and discuss the system performance and complexity of these three different weight algorithms：Singular Value Decomposition(SVD) , Maximum Norm Combining(MNC), Maximum Ration Transmit(MRT), and Improved Maximum Ration Transmit(IMRT). We simulate the system performance over Rayleigh fading channel, and compare these algorithms bit error rate(BER) at the same signal to noise ratio(SNR). In the case which the number of transmit antenna is small, four kinds of algorithms for performance is very close, and the complexity of the MRT algorithm is very low, but performance is reduced when the number of receiver antenna increase. The MNC method compares with SVD method that offers a good tradeoff between complexity and performance when the transmit antenna is increase.[[note]]碩