The classical approach for solving magnetic structures is the "trial and error" (TE) method. The reason is that many studies have been performed on relatively simple systems and the fact that the dominant isotropic exchange interactions, in absence of frustration, favour simple collinear structures. The improvement of the resolution of neutron powder diffractometers has allowed handling complex incommensurate magnetic systems. For solving complex magnetic structures the TE method is unable to provide appropriate results and other methods have to be used. In this communication the techniques for magnetic structure determination from neutron powder diffraction (NPD) data are reviewed. In the general case the magnetic moment of an atom in the crystal is given as a Fourier series. The Fourier coefficients, S-w, are complex vectors constituting the "unknowns" to be determined. These vectors define the magnetic structure and they correspond to the "atom positions" of an unknown crystal structure. The steps for solving magnetic structures from NPD are the following: i) Search for the propagation vector(s) {k}. The set {k} provides the. translation symmetry of the spin configuration. ii) Symmetry analysis is needed to find the smallest set of free parameters. In general the vectors SW are linear combinations of the basis functions of the irreducible representations of the wave vector group G(k). iii) Use an appropriate method for determining the coefficients of the above linear combinations. This implies an evaluation of the observed versus calculated intensity of the magnetic reflections. The use of a Patterson-like function or direct methods is not appropriate for magnetic structure determination due to the small number of useful reflections. The simulated annealing technique is more efficient in this field. We have improved and extended this method to the case of incommensurate magnetic structures.