-

Adaptive FIR digital filters are vastly used for system identification, equalizers, active noise
conceling , acoustic echo conceling, speech coding, etc. In applications which noise measurment or interference is large, algorithms such as LMS, NLMS, RLS in time domain or FDAF, FBAF, in
frequency domain suffer from unstability problem. In this paper, a new algorithm using the steepest descen t method is introduced in which
auto-correlation and cross-correlation functions
are used to adapt the coefficients of adaptive FIR digital filter.
This, algorithm specifically is suitable for the
m-dependent process and it can be used for applications such as the able double - talk echo canceling. To illustrate the performance of CLMS
algorithm, computer simulations are represented.