Collimation Lens Design using AI-GA Technique for Gaussian Radiators with Arbitrary Aperture Field Distribution

Abstract

The genetic algorithm (GA) is applied to the optimization of a sub-millimeter wave horn-lens collimation equipment operating at 150 GHz, and measurement is carried out to validate the performance of the optimized geometry. The aperture-integration (AI) method is employed to formulate the fitness function of GA in dealing with complex field distribution from an arbitrary radiation source. A diagonal horn and the auxiliary lens are fabricated according to the GA-optimized geometry, and their radiation properties are measured employing a BWO (Backward-wave Oscillator) with a golay cell terahertz detector. The measurement indicates that the gain and near-field beam radius conform to the values predicted by GA.