Skip to Main Content
A modified adaptive processor using an analytic signal with complex weights is presented to reduce the dimensions of input vector and weight vector in an adaptive array under directional constraint. Therefore, the rate of mean-squared convergence and misadjustment of the modified adaptive array by the constrained least-mean-square (LMS) algorithm can be improved. Moreover, the convergence rate of the constrained Kalman algorithm for estimating optimum weight vector also can become twice that of Er and Cantoni's algorithm. Also, we prove that the proposed method obtains the same optimal weight vector as Er and Cantoni's method. Computer simulations demonstrate that the modified adaptive array results in a great improvement in the rate of mean-squared convergence and misadjustment.