论文部分内容阅读
该文将非比例选择、保证收敛且易于判断收敛的整体退火遗传算法应用于优化线阵方位估计性能。优化过程中,不仅优化阵元位置改善方位估计,而且将阵元个数作为优化变量,从而提供了更多的自由度来控制阵列性能。在满足空间谱估计精度的条件下,利用较少的阵元实现超分辨估计。优点是减少信号处理量的同时,简化了设备。仿真实验结果表明该方法收敛速度快,有极强的避免过早收敛及避免局部极值的全局优化能力。
In this paper, the global annealing genetic algorithm with non-proportional selection, guaranteed convergence and easy to judge convergence is applied to optimize the linear array azimuth estimation performance. In the optimization process, not only the azimuth estimation is improved, but also the number of array elements is used as an optimization variable, which provides more freedom to control array performance. Under the condition of satisfying the spatial spectrum estimation accuracy, the super-resolution estimation is realized by using fewer array elements. The advantage is to reduce the amount of signal processing at the same time, simplify the equipment. Simulation results show that this method has the advantages of fast convergence, strong global optimization ability to avoid premature convergence and avoid local extremum.