基于改进DE的非等间距线阵综合

Unequally Spaced Linear Array Synthesis Using Modified DE

针对非等间距直线天线阵列的方向图综合问题,由于系统非线性较强,提出一种改进差分进化(DE)算法。改进DE算法的变异算子引入精英个体以加速收敛,精英个体的产生采用了随机化思想以保持种群多样性。此外,改进DE算法采用参数自适应策略来避免控制参数的调整。将改进DE算法应用到几个具有不同阵元间距约束的位置-相位综合实例中。综合的目标是降低旁瓣电平和控制给定方向的零陷。仿真结果表明,与经典的DE算法相比,改进DE算法具有较快的收敛速度,能够得到更好的综合结果。

A modified differential evolution （DE） algorithm was proposed for the synthesis of unequally spaced linear antenna arrays. The modified DE algorithm was used to adopt a new mutation operator in which elite individuals were introduced to accelerate the evolution. And elite individuals were generated by randomization to preserve population diversity. Moreover, the self - adaptive parameter control policy was employed to avoid tuning the parame- ters of the algorithm. Several position - phase synthesis problems for different restrictions on the distance between ad- jacent elements were studied with sidelobe level suppression and null control in specified directions. Synthesis results show that the proposed algorithm outperforms the classical DE in terms of convergence rate and the best values.