engleski

Efficient Methods for Design of Multirate Digital Filters

Multirate signal processing is widely used in various communication systems. In such applications, the most demanding operation is sampling rate conversion. A well-known sample rate converter employs the cascaded-integrator-comb (CIC) filter. However, its magnitude response often does not meet the requirement for selectivity. Popular techniques for improving the response are compensation and sharpening. In this work, we describe the design of compensated filters obtained by connecting an additional FIR filter whose coefficients are obtained by optimizing minimax error over the sum-of-powers-of-two space or obtained by using maximally flat error criterion. Furthermore, in the design of sharpened CIC filters, we consider the filters obtained by weighted least squares and weighted minimax method. The features of the presented methods are illustrated with the design of narrowband and wideband filters. The filters presented are suitable for application in narrowband and wideband software radio receivers.

multirate signal processing; communication systems; cascaded integrator-comb filter; CIC; compensation; sharpening

nije evidentirano