The Research On The Synthesis For Sparse Linear Antenna Arrays

The synthesis of linear sparse array antennas plays an important role in engineering applications such as satellite communications.Sparse array antennas can not only reduce cost and antenna volume,but also effectively reduce the sidelobe level and perform grating lobe suppression,improve the resolution of the antenna and simplify the feed network at the same time.It has been applied in many fields and received more and more attention.In recent years,the synthesis technology of non-uniform linear array antennas has tended tobe diversified.On the one hand,it is the traditional numerical analysis method such as Chebyshev and Taylor synthesis method,which has obvious limitations and can not solve the complex array antenna synthesis problem with multiple constraints.On the other hand,intelligent optimization algorithms are emerging with the development of computer technology.Such as genetic algorithms,particle swarm algorithms and other evolutionary algorithms.The intelligent algorithm can not only solve the problem of array element spacing under different restrictions and constraints,but also solve the problem of searching for the global optimal solution.In this paper,the synthesis of array antenna is realized through the improvement of the classical bat algorithm.Under the conditions of different array element spacing and array aperture constraints,the array element position of the linear symmetrical array antenna is optimized to obtain a lower peak sidelobe level.On this basis,this paper uses the convex optimization method under the framework of compressed sensing theory to synthesize the linear array antenna with as few elements as possible,and realizes the successful matching between the array synthesis pattern and the reference pattern.The main specific research work is as follows:(1)An improved bat algorithm(IBA)is proposed to solve the linear array synthesis problem with upper and lower bounds and aperture constraints,and the position of each element is optimized to obtain a lower peak sidelobe level.The improved bat algorithm embeds an adaptive local search strategy to enhance population diversity.Through numerical simulation,under different constraints of odd symmetric array,the optimal peak sidelobe levels of 17 and 37 array elements are 19.899 d B and 21.097 d B,respectively.In an even symmetric array,the peak sidelobe levels of 18 and 32 elements are-20.081 d B and-22.656 d B.Compared with other algorithms,the proposed method achieves a lower sidelobe level.(2)In this paper,a convex optimization method based on compressed sensing theory is proposed to reduce the number of array elements for the synthesis of sparse array antennas.Compressed sensing theory is used to sparse the linear uniform array antenna to solve the problem that the computing speed is too slow due to the excessive number of array elements.The original signal of compressed sensing is reconstructed,and the sequential convex optimization algorithm is used to obtain the matching element position,the number of elements,and the excitation amplitude corresponding to each element.The accuracy of the proposed method is verified by simulation,and it has the advantage of easy realization.The requirement of using fewer array elements to achieve the required array radiation pattern is achieved.