Maximum Correntropy Based Spectral Redshift Estimation for Spectral Redshift Navigation

Abstract

This paper presents a maximum correntropy method of redshift estimation for spectral redshift navigation (SRN). It converts the problem of spectral redshift estimation to a problem of maximum correntropy. A wavelet transform technique is established to denoise spectral signals and fit spectral continuous portions for spectral line extraction and further construction of a candidate redshift set. Subsequently, the joint correntropy is established by using the correntropy based on the redshift predicted from the SRN system equations to correct the correntropy based on the candidate redshift set. According to the maximum correntropy criterion, the spectral redshift estimation is acquired as the candidate redshift with the maximum joint correntropy. Simulations and comparative analysis demonstrate that in addition to the possessed real-time performance, the proposed method can also improve the accuracy of redshift estimation, leading to the improved accuracy for spectral redshift navigation.