: 2014Äê10ÔÂ13ÈÕ£¨ÖÜÒ»£©ÉÏÎç9:00
: У±¾²¿¶«ÇøÏèÓ¢´óÂ¥808ÊÒ
½²×ù: Recent Progress In Empirical Mode Decomposition
¾Ñéģ̬·Ö»¯µÄ×îнøÕ¹
: Dr. Boqiang Huang(»Æ²©Ç¿)
Universitaet Paderborn, Germany
Ñݽ²Õß¼ò½é£ºDr. Boqiang Huang received the B.S/M.S degree in 2004/2007 from Math&Physics/E.E. department, Yunnan Unviersity, and then received the Ph.D degree in 2010 from E.E. department, Fudan Unviersity. From 2010 to 2012 he was an Alexander von Humboldt PostDoc Fellow in Germany. He is now with the Institut fuer Mathmatics, Universitaet Paderborn, Germany. His recent research areas are mathematical signal processing, data analysis, data decomposition, optimization, pattern recognition, and data compression. He also interests in the (1-/multi- dimensional) real-world data analysis in many disciplines, e.g. applications to biomedicine, meteorology, hydrology, and economics.
½²×ùÌáÒª£º
The empirical mode decomposition (EMD) was firstly designed for nonlinear and/or nonstationary signal analysis. Combined with the Hilbert transform, the Hilbert-Huang-Transform (HHT) provides a finer time-frequency spectrum of a given signal compared with other well-known methods such as the Fourier transform, the wavelet transform, or the Wigner-Ville transform. Moreover, the EMD does not require any pre-defined basis. It decomposes a given data into a finite number of intrinsic mode functions (IMFs) and a monotonic trend. This talk will give a substantial introduction of the EMD and its modifications to different data or applications.
Ó½Ó¿í´óÀÏʦºÍѧÉú²ÎÓ룡
