Development of tri-band RF filters using evolutionary strategy

This paper is regarding the development of a tri-band bandpass filter based on T-shape resonators for multi-standard wireless networks. The filter dimensions have been optimized by using an evolutionary strategy algorithm in order to obtain three resonance frequencies and two transmission zeros. Our algorithm is a friendly and computational intelligence approach based on biological inspirations from the parents and offsprings relations for multi-object problems. Experiments results and full-wave simulations validate the proposed strategy and demonstrate its efficiency. The filter prototype provides three bands at 0.75, 1.67 and 2.60 GHz with a maximum insertion loss of 0.99 dB.