Tensor decomposition based R-dimensional matrix pencil method

Fuxi Wen, Wee Peng Tay

Abstract: In this paper, we extend the standard matrix pencil (MP) method to R-dimensional (R-D) tensor based MP. Higher-order singular value decomposition (HOSVD) is used to obtain the signal subspace. Performance of tensor based MP method is evaluated by computer simulations. Comparing with the conventional matrix based MP methods, better performance is obtained for tensor based R-D MP methods by exploiting the structure of the measurement data. Furthermore, it is straightforward to extend the proposed R-D tensor MP to other MP type methods, such as R-D unitary tensor MP, R-D beamspace tensor MP.