A HYBRID OPTIMIZATION FOR PATTERN SYNTHESIS OF LARGE ANTENNA ARRAYS

The pattern synthesis for large antenna arrays has drawn significant attention because of its wide applications.This paper introduces a hybrid approach for the fast pencil beam pattern synthesis of the large non-uniform linear or planar array, which can significantly reduce the computational cost, the number of antenna in the array, the minimum sidelobe level and the null control.The proposed method has an iterative scheme which is composed of the nonuniform Fourier transform (NUFFT) and the global optimization method to minimize the peak sidelobe level and control the null. The NUFFT is utilized to determine excitation magnitudes for a fixed positions non-uniform array. The global optimization is used tofind the optimal postions lead thepeak sidelobe level minimum alternatively.The lower excitations can be deleted due to yielding less performance on sidelobe level, which is calledthe array removal strategy.Compare with conventional methods,the simulations on synthetic models show thata minimum sidelobe level and null control can be obtained in processing sparse linear and concentric circular antenna arrays more efficiently.