【摘要】 研究了基于双传感器观测数据的二阶统计量和高阶统计量的复模糊函数到达时差／频差(TDOA／FDOA)同时估计算法 CAF-SOS、CAF-HOS。研究发现,CAF-SOS 在通常情况下估计精度较高,但不能对付色噪声,即对传感器噪声间的相关性比较敏感。而 CAF-HOS 则可以对付服从高斯分布的色噪声,但付出的代价是数据收敛性能慢,计算负担较大。为了解决计算效率问题,进一步研究了一种基于二阶统计量的二维处理方法(SD-MOM),将搜索范围由平面降为2条直线,因而总体计算成本得到明显改善。但该方法的缺点是不能处理色噪声,为此提出了一种基于累积量运算的分维估计方法(SD-CUM)。更多还原

【Abstract】 The estimation algorithms of CAF-SOS and CAF-HOS for TDOA/FDOA based on complex ambiguity function of second-order statistics and high-order statistics of dual-sensor observation data are discussed.We find that the CAF-SOS method has higher estimate accuracy,but not for color noise,that is the method of CAF-SOS is sensitive for correlation between sensor noises.The method of CAF-HOS can meet color noise following Gauss distribution,but has slow data convergence performance and high calculation load.In order to improve calculation efficiency,we study the two-dimensional processing method of SD-MOM based on second-order statistics.The search range is reduced to two lines from plane,which improves the overall calculation cost. Due to the disadvantage of ineffectivity for color noise of the above method,we bring forward the method of SD-CUM based on cumulation operation.更多还原