A three-dimensional structural health monitoring (SHM) system based on multiscale cross-sample entropy (MSCE) is proposed. Ambient vibration signals on the roof of a structure are measured, and vertical MSCE analysis is applied to evaluate the vertical damage location by analyzing individual signals on different floors. Results are quantified using the vertical damage index (DI). The damage orientation of damaged floors is then detected by analyzing the biaxial signals in four directions of each damaged floor through the planar MSCE analysis. Results are physically quantified using the planar DI. According to the results obtained from the evaluation of the proposed system on a three-dimensional structure, the damage condition and elevation were reliably diagnosed. Moreover, average accuracy rates of 93% (vertical) and 91% (planar) were achieved through the proposed DI method. Therefore, structural damage can be reliably detected after major earthquakes.