An adaptive kernel width update method of correntropy for channel estimation

Abstract

The maximum correntropy criterion (MCC) is a robust adaptation criterion, which has been successfully applied in various signal processing applications. The kernel width plays an important role in MCC adaptive filtering and has significant influence on convergence rate and steady-state performance. In this paper, we develop an adaptive kernel width update method for correntropy. The proposed algorithm may achieve faster convergence speed and lower steady-state excess mean square error (EMSE) than the original MCC algorithm. More importantly, the new method involves no extra free parameters and is computationally very simple.