Cumulant based maximum likelihood classification for overlapped signals

Authors:

Abstract
A novel automatic modulation classification algorithm named cumulant‐based maximum likelihood classification (CMLC) is proposed for overlapped sources. The sample estimate of cumulant is utilised for classification and classification decision is made by maximising the asymptotic distribution function of the cumulant. Simulation results prove the superior performance of CMLC over existing algorithms.
Funding Information
  • National Natural Science Foundation of China (61227801)

This publication has 4 references indexed in Scilit: