最大似然（ML）准则和最大后验概率（MAP）准则，是一种双隐层反向传播神经网络，基于分段非线性权重值的Pso算法，采用热核构造权重，合成孔径雷达（SAR)目标成像仿真，最终的权值矩阵就是滤波器的系数，DC-DC部分采用定功率单环控制，最小二乘回归分析算法。- Maximum Likelihood (ML) criteria and maximum a posteriori (MAP) criterion, Is a two hidden layer back propagation neural network, Based on piecewise nonlinear weight value Pso algorithm, Thermonuclear using weighting factors Synthetic Aperture Radar (SAR) imaging simulation target, The final weight matrix is ??the filter coefficient, DC-DC power single-part set-loop control, Least-squares regression analysis algorithm.