Joint Diagonalization Algorithm for Harmonic Retrieval

Roy R, Paulraj A, and Kailath T. ESPRIT-A subspacerotation approach to estimation of parameters of cisoids innoise[J].IEEE Trans. on Acoust., Speech, Signal Processing.1986, 34(10):1340-1342[2]Roy R and Kailath T. ESPRIT estimation of signalparameters via rotational invariance techniques[J].IEEE Trans.on Acoust, Speech, Signal Processing.1989, 37(7):984-995[3]Yeredor A. Non-orthogonal joint diagonalization in theleast-squares sense with application in blind sourceseparation[J].IEEE Trans. on Signal Processing.2002, 50(7):1545-1553[4]Vollgraf R and Obermayer K. Quadratic optimization forsimultaneous matrix diagonalization[J].IEEE Trans. on SignalProcessing.2006, 54(9):3270-3278[5]Vanpoucke F, Moonen M, and Berthoumieu Y. An efficientsubspace algorithm for 2-D harmonic retrieval. Proc. IEEEICASSP, Adelaide, Australia, 1994: 461-464.[6]Zhou Yi, Feng Dazheng, and Liu Jianqiang. A novelalgorithm for two-dimensional frequency estimation [J].Signal Processing.2007, 87(1):1-12[7]Feng Dazheng, Zhang Xianda, and Bao Zheng. A neuralnetwork learning for adaptively extracting cross-correlationfeatures between two high-dimension data streams[J].IEEETrans. on Neural Networks.2004, 15(6):1541-1553[8]Ziehe A, Laskov P, and Nolte G. A fast algorithm for jointdiagnolization with non-orthogonal transformation and itsapplication to blind signal separation. J Mach. Learn. Res.,2004, 5(7): 777-800.[9]Golub G H and Loan C F V. Matrix Computation. Baltimore,MD: Johns Hopkins Univ. Press, 1989, Chapter 10.