A Novel Parameter Estimation Algorithm for UAV by the Combined CVMD and TMSST Technique

Most existing parameter estimations for small unmanned aerial vehicles (UAV) have been transmuted into the extraction of micro-Doppler features. Linear time-frequency analysis methods represented by Short-Time Fourier Transform (STFT) and Wavelet Transform (WT) fall short in the flexibility of adaptation to the time window widths. Meanwhile, nonlinear approaches of time-frequency analysis, such as the Pseudo Wigner-Ville Distribution (PWVD), are verified arduous to eliminate the influence of cross terms. These methods apparently possess innate shortcomings. In this paper, we make the first attempt to introduce a combination algorithm to estimate the frequency of UAV rotors. Firstly, we employ Complex Variational Mode Decomposition (CVMD) to decompose the echo signals. And then, we adopt Time-Reassigned Multisynchrosqueezing Transform (TMSST) to process high-frequency components. Secondly, we use Time-domain Fast Fourier Transform (FFT) to obtain the CFD spectrum. Finally, we use Singular Value Decomposition (SVD) to estimate the rotor speed. The simulation and the experimental data show that our proposed combination algorithm not only improves the time-frequency energy aggregation, but improves the accuracy of blades rotation estimation.