【摘要】：定正則化因子的改進(jìn)多帶結(jié)構(gòu)子帶自適應(yīng)濾波(IMSAF)算法在取得收斂速度快和穩(wěn)態(tài)失調(diào)誤差小之間存在沖突.根據(jù)系統(tǒng)噪聲抵消原理,設(shè)定子帶后驗(yàn)誤差功率等于子帶噪聲功率,本文提出了變正則化矩陣的IMSAF算法來解決這一問題.仿真結(jié)果證明,所提算法可以同時(shí)達(dá)到收斂速度快、穩(wěn)態(tài)失調(diào)誤差小以及追蹤速度快等優(yōu)勢.
[Abstract]:The modified multiband subband adaptive filtering (IMSAF) algorithm with definite regularization factor has a conflict between fast convergence and small steady-state misalignment error. According to the principle of system noise cancellation, the posteriori error power of the subband is equal to the noise power of the subband. In this paper, a variable regularization matrix IMSAF algorithm is proposed to solve this problem. The simulation results show that the proposed algorithm can achieve the advantages of high convergence rate, small steady-state misalignment error and fast tracking speed.
【作者單位】： 中國科學(xué)院噪聲與振動(dòng)重點(diǎn)實(shí)驗(yàn)室(聲學(xué)研究所);中國科學(xué)院大學(xué);
【基金】：國家自然科學(xué)基金(No.61501449,No.11404367) 中國科學(xué)院先導(dǎo)專項(xiàng)項(xiàng)目(No.XDA06040501)
【分類號(hào)】：TN713;TN911.7
，

本文編號(hào)：2241074