采用互高阶累积量 (互四阶累积量 )的一维对角切片为统计量 ,证明了互高阶累积量可以有效地抑制非相关噪声和高斯噪声 ,并在建立互高阶累积量的Yule Walker方程的基础上 ,对离散系统 (AR过程 )的结构进行了分析。结果表明 ,该方法具有良好的谱估计的分辨率和谱估计的稳定性。与其它方法比较 ,该方法抗干扰性更强 ,其信噪比工作门限更低 ,有很好的应用前景 Using the one-dimensional diagonal slices of mutual high-order cumulants (mutual fourth-order cumulants) as the statistics, it is proved that the mutual high-order cumulants can effectively suppress the non-correlated noise and the Gaussian noise. Yule Walker equation, the structure of the discrete system (AR process) is analyzed. The results show that this method has good resolution of spectral estimation and stability of spectral estimation. Compared with other methods, this method is more robust to interference and has a lower signal to noise ratio working threshold and good application prospect