【摘要】 在建立互高阶累积量的Yule Walker方程的基础上 ,通过特征分解 ,建立了噪声矢量空间和信号矢量空间 ,提出了在非相关噪声和相关高斯噪声下 ,正弦参数估计的互高阶谱最小范数法。通过仿真计算可知 ,该方法信噪比工作门限低 ,且具有较高的谱估计分辨率更多还原

【Abstract】 Yule Walkers’ equation for cross high order accumulation is formulated as a basis in this paper,and by means of characteristic decomposition,noise vector space and signavector space are established.Cross high order spectrum minimal norm method is proposed for estimating sinusoidal parameters on background of non correlative noises and correlative gauss noises.Simulation results show that the novel method is with lower signal/noise ratio requirement and with higher spectrum identifying ability,therefore more suitable for practice.更多还原