LMS和歸一化LMS算法收斂門限與步長的確定
Determining of convergent threshold and step-size for lms and normalized lms algorithm
-
摘要: 從LMS算法失調(diào)量的準確表達式出發(fā)，根據(jù)輸入信號特征值分布重新研究了LMS，歸一化LMS(Normalized LMS，NLMS)算法收斂的必要條件,推導(dǎo)出LMS和NLMS 算法收斂的步長門限，并分析了輸入信號特征值分布、濾波器階數(shù)對算法收斂步長門限的影響，推導(dǎo)出滿足性能失調(diào)下步長的自適應(yīng)計算公式，減小了應(yīng)用 LMS，NLMS算法時步長選取的盲目性，與已有的算法相比，具有計算簡單、實用、自適應(yīng)性能強，同時可獲得滿意失調(diào)量的特點，計算機模擬結(jié)果表明該方法的正確性。
-
關(guān)鍵詞:
- LMS; 歸一化LMS; 穩(wěn)態(tài)失調(diào); 步長; 門限
Abstract: Based on eigenvalue distributing of input signal, the convergent necessary, condi-tions for LMS and Normalized LMS(NLMS) algorithm are again researched through the accu-rate analysis of the misadjustment. To avoid blindness for applying LMS and NLMS algorithm, the convergent threshold and the simple adaptive calculating formula for the step-size of them are proposed. The influence of eigenvalue distributing of input signal and order of filter on the threshold of step-size is also analyzed. Compared with an existing algorithm, the characters of lower computational complexity, practicality and stronger adaptive are shown and the satisfied misadjustment is achieved by adopting the presented method for calculating the step-size. -
B. Windows, Adaptive Signal Processing[M], NJ: Prentice-Hall Inc, 1985, Charter 3.C.F.N. Cowan, Adaptive Filters[M], NJ: Prentice-Hall Inc, 1985, Charter 4.[2]Jianfeng Liu, A novel adaptation scheme in the NLMS algorithm for echo cancellation[J], IEEE Signal Processing Letters, 2001, 8(1), 20-22.[3]裴炳南,LMS算法的收斂與步長選取[J],通信學(xué)報,1994,15(5),106-111.[4]曾召華,韋力,劉少亭,一種改進的瞬變步長LMS自適應(yīng)濾波算法[J],電子科學(xué)學(xué)刊,1998,20(4),566-569.[5]R.H. Kwong, E. W. Johnston, A variable step size LMS algorithm[J], IEEE Trans. on Signal Processing, 1992, 40(7), 1633-1642.[6]K. Mayyas, T. Aboulnasr, A robust variable step size LMS-type algorithm: analysis and simulations[J]. IEEE Trans. on Signal Processing, 1997, 45(3), 631-639. -
計量
- 文章訪問數(shù): 4096
- HTML全文瀏覽量: 186
- PDF下載量: 691
- 被引次數(shù): 0
下載:
