The occurrence of damage in a structure produces changes in its global dynamic characteristics such as its natural frequencies, mode shapes, modal dampings, modal participation factors, impulse response and frequency response functions. In this paper, a newly derived algorithm based on changes in power spectral density (PSD) is presented. The algorithm is used to detect damage, predict its location and assess the extent of damage in structures. The proposed method is based on only the measured data without the need for any modal identification. The method is described theoretically and applied to the experimental data, from a steel bridge model and bookshelf structure. Several damage scenarios were introduced to the members of the test structures. The method detected the damage, determined the exact location and monitored the increase in damage.