Abstract

Phased antenna arrays provide ultimate performance in areas where high directivity and electronic scanning are needed. That performance is achieved by involving large number of radiators as well as corresponding control units. As a result, such systems become bulky and heavy. In order to reduce the number of control units, elements are grouped in subarrays with one of the control units, such as time delay, put at the subarray level. The drawback of this approach is that if elements are grouped into subarrays producing repetitive patterns in the array structure, radiation pattern of such array will be affected by undesired grating lobes. To eliminate that effect, subarrays of irregular shapes, such as polyominoes, are used. Still, those structures are an object for optimization. This work aims at applying optimization techniques like genetic algorithm to the problem of finding optimal structures of phased antenna arrays composed of polyomino-shaped subarrays. For this purpose a new mathematical model, new algorithm and optimization methods are developed. Application of those techniques showed significant advances in radiation characteristics, in particular sidelobe level. Also new features were enabled, for example, multi-beam radiation pattern forming.