Title :
An algorithm-based fault-tolerant method for the 2-D LMS algorithm
Author :
Schmitz, Christopher D. ; Jenkins, W. Kenneth
Author_Institution :
Coordinated Sci. Lab., Illinois Univ., Urbana, IL, USA
fDate :
31 May-3 Jun 1998
Abstract :
An algorithm-based fault tolerant method termed the fault tolerant least-mean-squares (FTLMS) algorithm is extended from 1-D to 2-D. The algorithm enables the LMS to correct for a particular type of weight update failure called a stuck-at fault. The key to the algorithm is an overcomplete set of vectors used to transform the input. In developing the 2-D version, the properties of performance-enhancing transforms are investigated, leading to a transform structure that gives the algorithm a more desirable convergence rate. Various transforms are evaluated according to their reduction in eigenvalue ratio (EVR) for stationary inputs
Keywords :
correlation methods; eigenvalues and eigenfunctions; filtering theory; least mean squares methods; 2D LMS algorithm; algorithm-based fault-tolerant method; convergence rate; eigenvalue ratio; overcomplete vector set; performance-enhancing transforms; stationary inputs; stuck-at fault; transform structure; weight update failure; Autocorrelation; Convergence; Eigenvalues and eigenfunctions; Equations; Error correction; Fault tolerance; Filters; Least squares approximation; Transforms; Zinc;
Conference_Titel :
Circuits and Systems, 1998. ISCAS '98. Proceedings of the 1998 IEEE International Symposium on
Conference_Location :
Monterey, CA
Print_ISBN :
0-7803-4455-3
DOI :
10.1109/ISCAS.1998.694472
