论文部分内容阅读
对隐身飞机的雷达散射截面(RCS)统计建模时,传统方法通过直接计算RCS样本的统计特征估计模型参数,可能会产生较大的拟合误差。本文提出采用贝叶斯-蒙特卡罗(Bayesian-MCMC)方法提高起伏模型的参数估计精度,从而减小模型的拟合误差。首先将卡方分布模型和对数正态分布模型进行贝叶斯推导,得到其特征参数的后验估计表达式。然后采用MCMC算法构造后验分布的马尔可夫链,从而计算特征参数的估计值。最后通过比较2种方法的拟合曲线及其误差可知,本文方法适用于2种起伏模型,模型参数的估计误差比收敛误差门限值低1~2个数量级,2种分布模型的拟合精度均提高50%以上。
When modeling RCS of stealth aircraft, the traditional method can directly estimate the model parameters of the RCS samples, which may result in a large fitting error. In this paper, Bayesian-Monte Carlo method is proposed to improve the accuracy of the parameter estimation of the undulating model, so as to reduce the fitting error of the model. First, the Bayesian model is derived from the chi-square distribution model and the log-normal distribution model, and the posterior estimation formula of the characteristic parameters is obtained. Then the MCMC algorithm is used to construct the post-test distribution Markov chain to calculate the estimated characteristic parameters. Finally, by comparing the fitting curves and their errors of the two methods, we can see that the proposed method is suitable for two kinds of undulations, and the estimation error of the model parameters is 1-2 orders of magnitude lower than the convergence error threshold. The fitting accuracy of the two distribution models Are increased by 50% or more.