由於日常生活中有許多應用會使用到聲學回音消除（Acoustic Echo Cancellation, AEC）的技術，因此本論文欲提出一聲學回音消除之系統，將此系統建構在一線性適應性濾波器（Adaptive Filter）之上。近年來有許多學者提出適應性濾波器之各種改良方式，包含比例式正規化最小均方（Proportionate NLMS, PNLMS）、改進的比例式正規化最小均方（Improved PNLMS, IPNLMS），可變步長（Variable Step Size, VSS）以及頻域（Frequency Domain）形式的適應性濾波器。因此本論文主張將以上方法結合成單一系統，重新推導方程式並且將之應用在聲學回音消除中。從實驗結果中可以發現，本論文所提出之方法有效於估測環境中之回音路徑（Echo Path），除了有較佳的收斂結果外，還有更少的執行時間。;Due to the acoustic echo cancellation(AEC) technology using in our daily life, this thesis wants to propose a system of AEC, which is based on a linear adaptive filter. In recent years, many scholars have proposed various ways to improve the adaptive filter, such as proportionate normalized least mean square(PNLMS), improved proportionate normalized least mean square(IPNLMS), variable step size(VSS) and frequency domain adaptive filter(FDAF). Therefore, this thesis combine the above method into a single system, and derive the formula to implement the AEC. The experiment results show that the method proposed in this thesis is effective in estimating the echo path, and there is less execute time besides the better convergence result.
