Abstract: In order to control the active noise existing widespreadly in the electronic and communication systems,an IIR digital filter s realization based on fast array algorithm is proposed in this paper.The concrete implementation of IIR filter is based on the deep analysis for the implementation by the traditional RLS algorithm.Through combination the transfer function of secondary path with the regression vector ψf(n) and considering the data structure of the regression vector ψf(n),the regression vector is splitted into the samples of the previous outputs of the filter ψ1f(n) and the samples of the reference signal ψ2f(n).By means of the norm preserving mapping and the inner-product matrix form,the array forms used to filter's coefficient updating can be obtained eventually.The computing complexity for the two algorithms is also compared.The simulation results show that the fast array algorithm presented in this paper not only reduces the computing complexity,but also for the cancellation control for active noise,its convergence speed with respect to the convergence speed of traditional RLS algorithm is fast about 46%.