<p>Structural health monitoring (SHM) plays a vital role in ensuring the safety and longevity of civil structures. Recent advances emphasize the need for techniques that can accurately detect damage through dynamic response indicators rather than relying solely on visual inspections. Among these indicators, natural frequencies have proven to be particularly effective. However, most existing methods require multiple sensors or prior knowledge of the structure. In this context, we propose a novel approach for identifying natural frequencies and assessing damage using the synchrosqueezed wavelet transform (SWT), capable of delivering high accuracy with minimal instrumentation. Modal properties, such as natural frequencies, serve as indicators that can be used to evaluate structural health. By monitoring the natural frequencies over time, it is possible to identify damage occurrence through variations in their values. Our approach is based on SWT with a newly proposed analytical mother wavelet and consists of three stages: Gaussian smoothing, SWT, and spline interpolation. To validate the effectiveness of the method, accuracy in natural frequency identification and sensor minimization were used as evaluation metrics. The method was applied to two benchmark structures: a reduced-scale laboratory model (Benchmark Phase I-IASC-ASCE) and the full-scale Tianjin Yonghe bridge. The results demonstrated that the proposed approach is robust and accurate, capable of blindly identifying all natural frequencies in both structures using a single sensor.</p>