Optimized Synthesis of Self-Equalized Microwave Filters

In this paper, a new methodology for synthesizing self-equalized filters is introduced. The methodology consists of optimizing the polynomial transfer and reflection functions considering amplitude and phase constraints simultaneously. The algorithm gives access to a more general class of functions than generalized Chebyshev (also called quasi-elliptic) functions. Two self-equalized filters are designed using this new methodology, and their performances are compared to general Chebyshev filters in order to put into evidence the interest of the proposed optimization algorithm. The first design leads to reduced group-delay variations, and the second one leads to a simplification of the hardware, reducing the number of resonators.