一种基于高阶累积量的自适应时延估计新方法

A New Adaptive Time Delay Estimation Method Based on Higher Order Cumulants

讨论了存在于相关高斯或对称分布噪声中的非高斯信号的时延估计问题。可通过采用基于高阶统计量的自适应方法估计出滤波器的权系数 ,根据权系数的最大值求得时延估计。为了能得到非整数采样间隔的时延 ,常用SINC函数对滤波器权系数进行插值。但为了得到精确的时延估计 ,该法需要足够长的滤波器权系数。提出了一种可在相关高斯噪声中 ,只要较少的滤波器权长就能对非高斯信号的时延进行估计的新方法 ,称为最小权值误差平方和法。它对整数或非整数采样间隔时延均能进行估计。

The problem of estimating the difference in arrival time of a non Gaussian signal corrupted by Gaussian or symmetric probability density function noise is considered. With the adaptive algorithm to estimate the maximum coefficient of the filter, the estimation of time delay can be achieved. In the case that time delay is not an integer multiple of the sampling interval, the filter weights with a sinc function are usually incorporated to obtain the time delay. However, the sinc interpolator requires a sufficiently long filter for accurate delay estimation. In this paper, a new method (called“minimum sum of squares of weighted error”) is proposed to realize the optimum delay estimation even with a short filter in the domain of higher order cumulants. The results of the theoretical analysis and simulation are presented to illustrate the effectiveness of the new method.