Quotient Signal Estimation
Domenico Napoletani, Carlos A. Berenstein, Perinkulam S. Krishnaprasad, Daniele Carlo Struppa · Birkhäuser Boston eBooks · 2008
In this paper, we propose a method for blind signal decomposition that does not require the independence or stationarity of the sources. We define suitable quotients of linear combinations of the images of the mixtures in a given frame and we show experimentally that such quotients can be used to recursively extract three sources from only two measurements. A general strategy to extract more than three sources from two measurements is proposed. These keywords were added by machine and not by the authors. This process is experimental and the keywords may be updated as the learning algorithm improves.