Pablo Soto Pacheco

Fast automated design of waveguide filters using aggressive space mapping with a new segmentation strategy and a hybrid optimization algorithm

Waveguide filters are key elements present in many microwave and millimeter-wave communication systems. In recent times, ever-increasing efforts are being devoted to the development of automated computer-aided design (CAD) tools of such devices. In this paper, a novel CAD tool based on modal analysis methods, which improves the efficiency and robustness of the classical aggressive space-mapping technique, is presented for those purposes. The use of a new segmentation strategy and the hybridization of a specific combination of several well-known optimization algorithms is proposed. The CAD tool has been successfully validated with the practical design of several H-plane coupled cavity filters in a rectangular waveguide for space and communication applications. A filter prototype for local multipoint distribution systems operating at Ka-band, and two tunable H-plane filters with tuning posts operating at 11 and 13 GHz, have been successfully designed, manufactured, and measured.

Practical Design of Filters Using EBG Waveguides Periodically Loaded with Metal Ridges

The dispersion diagram of infinite periodic structures is useful for the practical design of waveguide filters. Starting from the dispersion diagram of a unit cell, it is possible to generate a finite structure with very similar pass- and stop-bands (gaps). However, truncation of the infinite periodic structure degrades the pass-band performance. In this paper, these impairments are overcome by means of suitable waveguide tapers matching the impedance of the periodic structure to the access ports. As a result, the design of practical low-pass filters, derived from passive structures based on Electromagnetic Band-Gap (EBG) waveguides periodically loaded with metal ridges, is successfully addressed. According to this procedure, a five-order and an eight-order EBG low-pass filters are obtained after an optimization step. Measurements of a manufactured prototype fully validate the proposed approach.