Waveform-preserving blind estimation of multiple sources
Lang Tong, Victor C. Soon, Y. Inouye, Yuxuan Huang, R. Liu · 2002
Some fundamental theoretical results in waveform-preserving blind estimation of multiple sources (estimation of vector source signals from observations without knowledge of the characteristics of the transmission channel) are presented. It is shown that the central issue for blind signal estimation and blind channel identification is the identifiability of the transmission channel. Necessary and sufficient conditions of blind channel identifiability are given for the model of independent and uncorrelated sources. The results obtained serve as the foundation for the development of a waveform-preserving blind signal estimation algorithm.>