A new design procedure for digital filters is presented. It stems from the well-known eigenfilter design. In contrast to the classical method, the new one is more stable and does not require a reference point. A special feature is that the method is closer to the least-squares method than the conventional eigenfilter design. It is important to point out that the new method allows linear constraints to be incorporated. Also, a new adventage lies in the fact that the proposed method can be extended to 2-D linear phase filters design. This paper represents a masterpiece in filter design. The main ideas of the method are clearly presented, the derivation of the results are well conducted, and the numerical examples fully illustrate the advantages of the new procedure.
Reviewer: D.Stanomir (Bucureşti)