Optimization Research Of Sparse And Thinned Antenna Array Based On Cuckoo Search

Antenna array synthesis is the inverse problem of antenna array analysis.The purpose of antenna array beampattern synthesis is to determine the array element layouts and the array element excitations which produces a radiation pattern that is closest to the desired pattern.With the development of computer technology,intelligent optimization algorithms are widely used in the area of antenna design,because of its robust,fitting for mud-object and without requiring for differentiability.There are two intelligent optimization algorithms are discussed in this paper-Particle Swarm Optimization Algorithm(PSO)and Cuckoo Search(CS),and the applications of these algorithms in synthesis antenna.PSO algorithm is a newly evolutionary algorithm by simulating bees,fishes and other animals foraging behavior.The characteristics of the algorithm are:information sharing between group and individual behavior is influenced by group behavior.The CS algorithm is a self-heuristic search algorithm based on the interesting breeding behavior of the cuckoo.The efficiency of the algorithm lies in the close coordination of the Levy flight random walk and the preferred random walk.During the optimization process,frequent short-range searches and occasional long-distance jumps are alternately performed.Firstly,the theory of PSO algorithm and CS algorithm is introduced,and its validity is verified by the test function.The low side-lobe pattern and the flat-top pattern are optimized by using the PSO algorithm and CS algorithm for the symmetric 20-elements linear array.Fully understanding the optimization mechanism and the superiority of the algorithm by comparing the result with the Chebyshev and Woodyard results.Secondly,considering the minimum constraint condition of elements spacing,for the symmetric 20-elements linear array with array aperture 6?,optimizing the position of the elements to obtain the lower side-lobe pattern.Introducing the sparse array model that combines the quantitative amplitude and position of the elements,and taking into account the amplitude DRR value restrictions,the side-lobe level of the optimized pattern is significantly lower than that of the array elements' position optimized only.Finally,for the rectangular planar array with aperture 2L × 2L = 9.5? × 9.5? and 20 × 20 elements,the binary cuckoo search algorithm is used to get the ideal pattern.In this paper,the sparse planar array models which elements with different spacing is proposed.Two models can reduce the side-lobe level to a certain extent.Theoretically,array elements number,array aperture,and elements spacing constraint conditions are the same,the thinned plane array is desirable to obtain a lower side-lobe level than the sparse plane array because of the greater degree of freedom of the position.