Sparse Array Design Using the Genetic Algorithm for Optimizing the Radiation Pattern of Linear Arrays

Abstract

The genetic algorithm was used to obtain sparse arrays with 8, 16, 32 and 64 transducers in a possible aperture size of 128 elements. The optimization was made by minimizing a fitness function that takes into account the mainlobe width and sidelobe levels of the array radiation pattern. The sparse arrays were used to image 11 twisted wires in a medical phantom. The sparse arrays were able to generate images with resolution comparable to a 128 elements array and reduced artifacts in comparison to arrays with the same number of elements.