在非理想環境下用於多輸入多輸出雷達之可適性波束成型技術

Abstract

本篇論文將討論多輸入多輸出雷達(Multi-input Multi-output Radars)架構上的波束成型器設計，針對不同的指引向量誤差，像是角度誤差、相加型指引向量誤差、已知天線耦合現象、元件位置擾動以及局部散射，我們利用實驗室所設計的極小極大準則估計訊號真實的方向向量，配合現有文獻上的干擾加雜訊自相關 (Interference-Plus-Noise Covariance , IPNC) 矩陣估計，可以有效的解決上述這些單一誤差甚至是多重誤差，並且效果會優於傳統的技術。若存在同調干擾訊號時，將我們的方法配合文獻上的分離式架構，可以有效改善同調所造成的影響。然而在同調干擾訊號環境、未知天線耦合現象和元件位置擾動下，我們改良了傳統的極小極大準則，將真實的方向向量拆成增益(Gain)和相位(Phase)分開估計，並且配合另一種現有文獻的干擾加雜訊自相關矩陣估計方法，用類似的概念套用到多輸入多輸出雷達架構上，設計出了可以解決此多重誤差的方法，效果也優於傳統的技術。因此透過估計真實的方向向量以及干擾加雜訊自相關矩陣，我們可以解決很多誤差問題，使得天線可以有效的在實際情況下接收到訊號，並且消除不必要的干擾以及雜訊。

This thesis presents the design of adaptive beamformer for MIMO (Multi-input Multi-output) radar systems in the presence of uncertain mismatches including angle error, additive-type vector error, mutual coupling, position error, and local scattering. We use a minimax algorithm to estimate the actual direction vector of the desired signal, and then develop a robust method based on the minimax algorithm in conjunction with an Interference-Plus-Noise Covariance (IPNC) matrix estimation. This method can deal with considered mismatches effectively and provide better beamforming performance than the existing methods. Moreover, the robust method can simply incorporate with a separate structure of MIMO radar systems to alleviate the performance degradation due to coherent signal sources. During the estimation of the direction vector of the desired signal, we have also found that estimating the gain and phase of the direction vector separately can further enhance the robust capabilities of the proposed method. Computer simulation results show that the proposed method is effective in dealing with the above mentioned mismatch problems for MIMO radar systems and provide better beamforming performance as compared to the existing methods.