论文部分内容阅读
针对海杂波K分布形状参数和尺度参数的估计问题,提出了基于分数阶矩和NM单纯形算法的参数估计,首先采用分数阶矩估计法作为形状参数和尺度参数的初始值,然后利用NM单纯形算法进行数值搜索获得K分布形状参数和尺度参数的估计值。与实测的X波段海杂波概率密度函数对比表明,K分布优于Gamma分布和Weibull分布,其估计值与实测值最相吻合,且具有最小的均方根误差。对于一维的海杂波来讲,该方法是一种计算效率适中的算法。
In order to solve the problem of the estimation of the shape and scale parameters of the K distribution in sea clutter, the parameter estimation based on fractional moment and NM simplex algorithm is proposed. Fractional moment estimation is adopted as the initial value of the shape parameter and the scale parameter. Simplex algorithm for numerical search to obtain K distribution shape parameters and scale parameters of the estimated value. Compared with the measured X-band sea clutter probability density function, it is found that the K distribution is superior to the Gamma distribution and Weibull distribution, and the estimated value is in good agreement with the measured value and has the smallest root mean square error. For one-dimensional sea clutter, this method is a computationally efficient algorithm.