A Sparse Representation Algorithm for the Mode Separation and SNR Improvement of Lamb Wave Signals

A matching pursuit method based on training an over-complete dictionary is investigated. By using the K-singular value decomposition (K-SVD) algorithm, an over-complete dictionary that describes the Lamb wave signals is trained. The results demonstrate that the method can effectively remove redundant information and separate multiple Lamb wave modes. The matching pursuit method with the K-SVD algorithm shows its efficiency for Lamb wave signal processing.