In assuring the safety of a structure, it is important to evaluate the state of structural members non-destructively by magneto-mechanical acoustic emission (MAE). MAE is useful in NDT, because it reflects minute changes of stress and microstructure of ferrous materials. We studied MAE signals of a pure iron using displacement and shear sensors, for the detection of the longitudinal and shear components of the signals, which were analyzed by wavelet transform. The mapping of wavelet coefficients in the time-frequency coordinate plane provides informative characterization of the signals. From the analysis, we found that the low (150 to 300 kHz) and high frequency (400 to 800 kHz) components are caused by different magnetization effects. The low frequency component has strong displacement sensor response coincident with the rate of magnetic induction, suggesting magnetization via the discontinuous motion of domain walls. The high frequency component has strong shear sensor response and is observed near magnetic saturation, suggesting rotational magnetization effect.