Correntropy of the Symmetric Stable Distribution and Its Application to the Time Delay Estimation
对称稳定分布的相关熵及其在时间延迟估计上的应用

Correntropy is a localized similarity measure between two scalar random variables. This paper presents the parametric representation of the symmetric α-stable (SαS) distribution’s correntropy. The equivalency of the maximum correntropy criterion and the minimum dispersion criterion is derived from the parametric representation for zero location SαS distributions. This result is used to propose the adaptive time delay estimation in SαS noise. Simulations show that the algorithm based on correntropy works better than the least mean square and the least mean p-norm approaches.