Microstrip bandpass filter design using split-path method and optimized curvature corrections

Bandpass filters with wide pass-band are an essential requirement in various equipments of satellite and defence communication sectors. Here a method of split-path interactions is proposed to approximately predict the resonant frequency and topology of bandpass filters which otherwise fall under the category of heuristic designs. Curved transmission lines are often required to make filter structures physically compact; however, curvature effects create errors in the theoretical (design) prediction of resonant or central frequencies for bandpass filter design. Earlier propositions on curvature corrections had been considerably precise, but recent design standards demand even higher accuracies. The prime feature of this work is the use of a meta-heuristic optimization (i.e. Particle Swarm Optimization) technique in curvature corrections for the first time which brings accuracies of over 99% in the corrected results. The split paths used in this design are suitably curved, with the proposed optimized curvature correction technique, so as to attain a compact size of the filter. The resulting filter has a low insertion loss of around -1.00dB and a sharp stopband cut-off. Fabrication was done on a FR4 microstrip substrate with a good agreement between measured and simulated results. Copyright (c) 2015 John Wiley & Sons, Ltd.