运用GA和FFT确定稀疏阵列的激励幅度

针对天线单元沿阵列中心非对称分布的稀疏直线阵列(单元从间距为半波长的规则栅格中稀疏),讨论了以激励幅度分布为决策变量,以最大相对旁瓣电平为优化目标的遗传算法(genetic algorithm,GA),运用个体的真值编码及其中间重组的交叉方法改进了遗传算法的收敛性能。由于阵元间距是栅格的整数倍,因此GA中凭借离散傅里叶变换,使适应度函数的计算可利用高效的FFT算法。仿真中对一个154阵元、孔径约100倍波长的非对称稀疏线阵的激励幅度进行优化,使其副瓣电平下降了1.36dB。两个仿真实例证实了算法的有效性。

Genetic algorithm and FFT for the synthesis of element weight coefficients in asymmetric thinned arrays

Elements of asymmetric linear thinned arrays are not bilaterally symmetric with respect to their centers,and thinned arrays are thinned from a uniformly linear array whose element spacing is half wavelength usually.An optimization approach based on genetic algorithm(GA) and aimed at array's performance improvement,that is,to lower the maximum RSLL(Relative Side Lobe Level) by designing element currents of an asymmetric linear thinned array' is presented.By real coding of chromosome and midterm recombination of individual the GA's convergence is effectively improved.Since the aperture quantization length is half wavelength and thus element spacing of thinned arrays is integral times of half wavelength,effective FFT is used to speed up the evaluation of fitness function in GAs.simulation for 154 element linear asymmetric thinned array with the aperture size approximate 100 wavelengths is inade and the maximum RSLL of the array reduces by 1.36dB.Two simulation results show the validity of the approach.