Adaptive Orthogonal Signal Decomposition Based On Empirical Mode Decomposition And Empirical Wavelet Transform
Empirical mode decomposition (EMD) and Empirical wavelet transform (EWT) are recently developed adaptive
signal processing tools. These techniques decompose a signal accordingly to its contained information. The main issue with
EMD is its lack of theory and in case of EWT a prior knowledge of the signal is required. IMF’s obtained as a result of
applying EMD are quasi-orthogonal .This paper suggest how to overcome these difficulties. For this Gram-Schmidt
orthogonalization procedure is applied to the IMF’s generated. The number of orthogonal components obtained determines
the number of modes for applying EWT to the same signal.