Title :
New Classes of LMS and LMF Adaptive Algorithms
Author :
Hosseini, Kianoush ; Montazeri, Aman ; Alikhanian, Hooman ; Kahaei, Mohammad H.
Author_Institution :
Electr. Eng. Dept., Iran Univ. of Sci. & Technol., Tehran
Abstract :
In this paper, some adaptive algorithms based on the Least Mean Square (LMS) and Least Mean Fourth (LMF) are proposed. The same approach as the Variable Step-Size LMS (VSLMS), Sign LMS (S-LMS) and Normalized LMF (XE- NLMF) algorithms is used. The convergence speed of these algorithms are compared to that of previous algorithms . It is shown that among the proposed algorithms Variable Step-size XE-NLMF (VS-XE-NLMF) and VSLMS+F have better convergence behavior than that of the previous algorithms.
Keywords :
adaptive filters; adaptive signal processing; convergence; least mean squares methods; adaptive algorithm convergence; adaptive filter theory; normalized least mean fourth algorithm; sign least mean square algorithm; variable step-size least mean square algorithm; Adaptive algorithm; Adaptive filters; Computer simulation; Convergence; Cost function; Gaussian noise; Iterative algorithms; Least squares approximation; System identification; Working environment noise; Adaptive Filter Theory; Least Mean Forth (LMF); Least Mean Square (LMS); System Identification;
Conference_Titel :
Information and Communication Technologies: From Theory to Applications, 2008. ICTTA 2008. 3rd International Conference on
Conference_Location :
Damascus
Print_ISBN :
978-1-4244-1751-3
Electronic_ISBN :
978-1-4244-1752-0
DOI :
10.1109/ICTTA.2008.4530045
