摘  要：被动探测的关键技术之一就是通过有效的信号处理方法提取出多接收器的时间差，而确定时差最基本的方法是进行信号相关以获得估计信息，相关算法虽然计算过程简单，但是算法对信号和噪声频谱特性要求比较高。基于希尔伯特变换的时延估计是广义相关时延估计方法的拓展，把相关函数的对称性由偶函数转化成奇函数，将相关法中对峰值的检测转化为对过零点的检测。通过对两种算法的误差比较分析，信号的截至频率越高，希尔伯特变换法的均方误差与基本相关法的均方误差相差越明显，估计误差越理想，证明Hilbert算法的优越性。

关键词：Hilbert；时延估计；误差分析；相关算法

文献标志码：A        文章编号：1674-5124（2015）12-0014-03

Error analysis of TDE based on correlation Hilbert transform

WANG Feng， LIU Pengyuan， LI Bing

（Foth Department，Ordnance Engineering College，Shijiazhuang 050003，China）

Abstract: Passive detection has the advantages of anti-reconnaissance， strong anti-jamming and concealment. It has become an important direction to raise the capacity of electronic countermeasures. One of the key technologies for passive detection is to extract the time lag through effective signal processing. Relevant method is the most basic information to estimate the time difference. It is the basic theory of all correlative time-delay estimation (TDE) algorithms. Calculation correlation method is simple. But satisfactory results cannot be achieved if the spectrum characteristics of signal and noise are not ideal. Time delay estimation based on Hilbert transform is the expansion of the generalized correlation time-delay estimation method which changes the correlation function from accidental symmetry into odd symmetry. Detecting correlation peak is converted into zero crossing detection. After comparing the error of the two kinds of methods， the higher the cutoff frequency is the more obvious when the differences between the mean squared errors of the two methods are larger. In other words， the estimate is more precise. 

Keywords: Hilbert； time-delay estimation； error analyzing； correlation algorithms