The dynamic properties and characteristics of engineering structures may change throughout their lives for a variety of reasons, often over time. In such engineering structures, for which the term "time-varying systems" can be used, it is important to identify time-varying dynamic parameters, including instantaneous modal frequency, as it can enable engineers to monitor the performance of structures. It also helps to diagnose structural damage and assess the condition of the structure in a cost-effective manner. In fact, tools that can correctly obtain characteristics similar to the instantaneous frequency in the analysis of such systems have particular importance. Tools for extracting such characteristics from dynamic responses are known as time-frequency domain analysis methods. These instantaneous characteristics can be used to identify time-varying parameters in these structures, such as time-varying stiffness and damping. In the current paper, two of the most powerful methods of time-frequency domain analysis, namely wavelet analysis, and analytical mode decomposition-Hilbert, are used to extract instantaneous characteristics. These two methods in extracting the characteristics are compared with each other, their superiority over each other in noise, and no noise conditions are examined. Then, by combining each of these two methods with existing algorithms in the parametric identification of multi-degree of freedom time-varying systems, new combinations are introduced that report performance improvements in the identification of time-variable parameters.