Research On Synthesis And Optimization Algorithms For Aperiodic Comformal Antenna Arrays

Taking airborne opportunistic array radar as the research background,considering the antenna elements of opportunistic array are attaching to arbitrary surfaces of three-dimensional carriers and the paramters such as excitation coefficients,positions and orientations of these elements are controllable,the synthesis and optimization problem for aperiodic conformal arrays has been studied in this paper.However,since the radiation pattern is a non-linear function of array element positions,the optimization problem of aperiodic array layout is a multi-dimensional nonlinear optimization problem,which has not been addressed efficiently to date.In this paper,several algorithms such as particle swarm optimization(PSO),convex optimization and orthogonal perturbation method(OPM)are utilized as the optimization tools to design different aperiodic arrays,including aperiodic linear array,aperiodic planar array,aperiodic circular array,aperiodic concial array and aperiodic cylinder array.The major research work and results of this paper are as follows:(1)The mathematic model of conformal antenna array is established firstly,the matrix form of array radiation pattern is given and the polarization pattern is also derived.The current research situation on pattern synthesis and array optimization is studied,with two algorithms: Chebyshev synthesis method and particle swarm optimization are analyzed.Simulation experiments are conducted to discuss the performances of these two algorithms.(2)The design of aperiodic conformal array with optimal directivity and sidelode and cross-polarization suppression is studied.As the optimization problem for conformal arrays with optimal directivity and low sidelode and low cross-polarization is a multi-objective optimization model,the hybrid algorithm based on multi-objective particle swarm optimization(MOPSO)and second-order cone programming(SOCP)is proposed.MOPSO is applied to optimize the array layout,while SOCP is used to find the optimal excitation coefficients for each layout,thus the optimal aperiodic conformal array solution can be obtained.Simulation results of aperiodic circular and conical arrays show that the proposed algorithm can synthesize aperiodic conformal arrays with better pattern characteristics compared with other normal evolutionary algorithms.(3)The geomeotry synthesis of arbitrary arrays based on orthogonal perturbation method(OPM)is studied.Considering that the problem of high time complexity and poor stability occurs when using evolutionary algorithms to deal with this nonlinear optimization problem,an efficient algorithm named OPM is presented.The radiation pattern is expressed as a linear combination of a set of orthogonal basis,and the array element layout is obtained by using an iterative procedure given the information of desired pattern and synthesized pattern.Various simulation examples demonstrate the effectiveness and efficiency of the proposed algorithm.(4)The orthogonal perturbation algorithm under non-uniform excitation is studied.An orthogonal perturbation methon with quantization excitation is given.Using the uniform array layout with non-uniform excitation as the initial point of iteration,the algorithm can better approximate the desired pattern and improve the convergence speed at the same time.On this basis,several convex models with different optimization constraints are constructed,and an optimization algorithm combining convex optimization and orthogonal perturbation is presented.The optimal element arrangement and its associated excitation strategy for aperiodic circular and cylinder arrays.