Skip to main content
. Author manuscript; available in PMC: 2017 Oct 1.
Published in final edited form as: IEEE J Sel Top Signal Process. 2016 Jul 27;10(7):1134–1149. doi: 10.1109/JSTSP.2016.2594945

Fig. 5. General inverse model for the linear MDM problem.

Fig. 5

The lower layer corresponds to one Vm × 1 observation of each input data stream xm, reflecting the different intrinsic dimensionality (Vm) of each dataset. The middle layer represents the Cm sources. The top layer establishes the K subspaces sk, which may be dataset-specific (k = K in the figure) or span through many datasets, illustrating the different compositions permitted.