Under the influence of an external magnetic field, the magnetic answer of a tested sample will reflect its content, its nature and its history [1]-[4]. The use of magnetic control for the nondestructive evaluation of structural steels has increased significantly recently [5]. The measurement of the Barkhausen noise provides the velocity of the magnetic domain walls. It indirectly reflects some precise information about the microstructural content of the tested materials. Because of its stochastic nature, the raw Barkhausen electromagnetic noise measured at the sensor output is extremely difficult to analyze. After a post-processing stage and a renormalization, it is possible to plot the energy hysteresis cycles from the Barkhausen noise measurement. These local cycles are much more stables reproductive and eventually easy to analyze [3][4]. They offer a controlled access to the evolution of the microstructure of the material tested (internal stresses, level of degradation ...) and make it an essential tool for the non-destructive evaluation of ferromagnetic steels. In this project, we propose a modeling of these magnetic Barkhausen noise energy cycles and indirectly the envelope of the highfrequency Barkhausen noise. By combining the multi-scale model [6] and the Jiles-Atherton one [7][8], we should be able to separate the rotation and the domain wall movements and reach an anhysteric curve made out from just the domain wall contribution. In this project many results of simulations and characterizations (classical cycles, MBNenergy cycles) will be compared in order to validate this theory