A multiple exchange algorithm for constrained design of FIR filters in the complex domain

We present a fast multiple exchange algorithm that designs FIR filters with magnitude and phase specifications subject to constraints on the error function. We use a constrained least squares criterion which minimizes error energy and imposes bounds on the magnitude of the error. We can trade error energy versus peak error, and complex least squares and complex Chebyshev filters result as special cases. We provide a Matlab program implementing the proposed algorithm. This program has proved to be efficient and reliable.