A Family of Normalized LMS Algorithms

被引:111
作者
Douglas, Scott C. [1 ]
机构
[1] Univ Utah, Dept Elect Engn, Salt Lake City, UT 84112 USA
关键词
Least mean square algorithms;
D O I
10.1109/97.295321
中图分类号
TM [电工技术]; TN [电子技术、通信技术];
学科分类号
0808 ; 0809 ;
摘要
In this letter, a derivation of the normalized LMS algorithm is generalized, resulting in a family of projection-like algorithms based on an L(p)-minimized filter coefficient change. The resulting algorithms include the simplified NLMS algorithm of Nagumo and Noda and an even simpler single-coefficient update algorithm based on the maximum absolute value datum of the input data vector. A complete derivation of the algorithm family is given, and simulations are performed to show the convergence behaviors of the algorithms.
引用
收藏
页码:49 / 51
页数:3
相关论文
共 6 条
[1]  
DOUGLAS SC, 1994, IEEE T ACOUST SP JUN, V42
[2]  
Goodwin GC, 1984, ADAPTIVE FILTERING P
[3]   A LEARNING METHOD FOR SYSTEM IDENTIFICATION [J].
NAGUMO, JI ;
NODA, A .
IEEE TRANSACTIONS ON AUTOMATIC CONTROL, 1967, AC12 (03) :282-&
[4]   FAST ALGORITHMS FOR RUNNING ORDERING AND MAX MIN CALCULATION [J].
PITAS, I .
IEEE TRANSACTIONS ON CIRCUITS AND SYSTEMS, 1989, 36 (06) :795-804
[5]  
Sung Ho Cho, 1991, IEEE Pacific Rim Conference on Communications, Computers and Signal Processing (Cat. No.91CH2954-6), P643, DOI 10.1109/PACRIM.1991.160821
[6]   CONVERGENCE AND PERFORMANCE ANALYSIS OF THE NORMALIZED LMS ALGORITHM WITH UNCORRELATED GAUSSIAN DATA [J].
TARRAB, M ;
FEUER, A .
IEEE TRANSACTIONS ON INFORMATION THEORY, 1988, 34 (04) :680-691