Synthesis Of Array Antennas Using Genetic Algorithm

With the development of wireless communication, the limited frequency resource declined rapidly. To optimize the resource allocation and improve the utilization ratio of resource is becoming a pursuing goal for scientists and engineers. Smart antenna technology occurred in such background, and pattern synthesis of antenna array is the core technology of smart antenna. It can utilize the resource effectively by adjusting the parameters of the antenna units such as the numbers, positions, feeding-back amplitudes and phases in different frequency circumstance. The comparative advantage of genetic algorithm is obvious because of its' ability of global rapid optimum.Genetic algorithm is a bionic algorithm which simulates the heredity and variation of creature. Based on the principle of "the survival of the fittest", it can find the optimal solution through the process of selecting, crossing and variation. Genetic algorithm has the advantages to settle large space, multi-parameter and no-linear problems, so it is applied in extensive field. In recent years genetic algorithm has application to analyze and design antenna.This thesis carried on a research in applying genetic algorithm to pattern synthesis of antenna array. Sidelobe suppression and null steering are researched. The control of low sidelob in ultra-low sidelobe antenna array and thinned array are discussed detailedly.Pattern synthesis of linear array and rectangular array using genetic algorithm is under the ideal condition in the thesis. Conclusions gained are as follows:Firstly, Genetic algorithm is the efficient method of pattern synthesis of antenna array.Genetic algorithm is an efficient intelligent bionic algorithm. It uses overall and local searching method to ensure the diversity of the colony, in order to avoid the earliness. In the thesis, array spacing,amplitude and phase of excitement in linear array and rectangular array are optimized using genetic algorithm. The result confirm with the design beacon which is the efficient control of sidelobe, ultra-low sidelobe and depth of the null steering in specified angle.Secondly, the performance of pattern synthesis using genetic algorithm is better than which using traditional method, and which is more evidence when problem required optimization is more complexity.Results show that pattern synthesized using genetic algorithm is more excellent than which of traditional method, and when the array amount is larger or the pattern function is more complicated, the superiority is more obvious.Thirdly, the pattern function of antenna array is transformed to discrete form.Through sampling to pattern function of antenna array, a discrete function is gain which is adapted to computer operating and genetic algorithm. Circulatory operation is replaced by matrixes' product, which advances efficiency of computing.