Through-the-wall radar imaging (TWRI) systems allow police, fire personnel, and defense forces to detect, identify, and track subjects inside buildings or under rubble. In this paper, the delay and sum (DAS) and range migration (RM) algorithms are compared as imaging techniques for a multiple input multiple output (MIMO) stepped-frequency (SF) radar system. These algorithms have been applied to analytical, simulated, and measured data both in the absence and in the presence of a wall between the antenna and the target. Both techniques were able to accurately reconstruct the position of targets behind a wall. The DAS presents a wider angle of nonambiguity while the RM is faster. An improvement of the DAS, in terms of accuracy in the target positioning, is achieved applying the Fermat's principle.