Title :
Nonrecursive filter training algorithm
Abstract :
Customization of adaptive filters is carried out with help of defined algorithms, on which the adaptation time and complexity of a setting system depends, therefore development of high-performance algorithms is the important theoretical and practical problem. In the report the filter training algorithm is considered on the basis of a minimum discrepancies method. The lower bound of a functional of minimizing is reduced, an exact estimation of algorithm rate of convergence is obtained.
Keywords :
adaptive filters; convergence of numerical methods; filtering theory; gradient methods; adaptation time; adaptive filter; algorithm rate of convergence; high-performance algorithms; minimum discrepancies method; nonrecursive filter; quickest descent algorithm; training algorithm; Adaptive filters; Adaptive signal processing; Adaptive systems; Communication channels; Communication switching; Convergence; Data communication; Intersymbol interference; Signal processing; Signal processing algorithms;
Conference_Titel :
CAD Systems in Microelectronics, 2003. CADSM 2003. Proceedings of the 7th International Conference. The Experience of Designing and Application of
Print_ISBN :
966-553-278-2
DOI :
10.1109/CADSM.2003.1255021
