Surrogate data generation technology using the SSA method for enhancing the effectiveness of signal spectral analysis

Spectral analysis of signals by Root-Min-Norm method, using modified SSA method

“…It can be seen that columns of the matrix are the vectors (segments) of size , denotes transposing, , . In the nonlinear dynamics the similar segmentation of time series is called embedding in the phase space [4].…”

“…The eigenvalue decomposition (EVD) of the CM of observations or singular value decomposition of the data matrix is the base of the modern spectral analysis methods. Furthermore, the methods of noise reduction, the methods of signal processing in communication systems with a multi-element antenna array on the transmitting and receiving side (MIMO-multiple inputmultiple output system), pattern recognition (recognition of signals with digital modulation or images) methods, methods of parameter estimation of compound signals also use EVD or SVD [1][2][3][4][5][6].…”

“…Efficiency of spectral analysis methods becomes worse in the conditions of the low signal-to-noise ratio, small samples. It can be improved using the preliminary signal processing (preprocessing) methods [2; [4][5].…”

“…The beamspace transformation, spatial smoothing, forward-backward averaging, noise reduction (signal filtering), Gram-Shmidt orthogonalizaiton, singular spectrum analysis (SSA) [1][2][3][4][5][6] are the examples of the preliminary processing methods that can be used with spectral analysis methods.…”

“…SSA is one of the methods of noise reduction in observation, such as the total least squares (TLS), surrogate data technology, wavelet transformation [4][5].…”