Feature extraction of laser micro-doppler signatures based on NMP

In order to accurately estimate the parameters of laser micro-Doppler signal, an innovative time-frequency analysis method, nonlinear matching pursuit (NMP), was employed. As there were great limitations in the strong background noise and weak modulation conditions, an approach designated as weighted average frequency algorithm (WAFA-NMP) was proposed to further improve the NMP. Weighted average frequencies and amplitudes were computed to determine the corresponding point on the time-frequency space. In addition, a sub-sampling method for the weak modulation problem was proposed based on WAFA-NMP. Simulation results showed that the proposed algorithm can achieve better performance compared with the Wigner-Ville (WV) distribution and smoothed pseudo Wigner-Ville (SPWV) distribution. The valuation accuracy is 96% on average, and the noise immunity can be low to -10 dB. ©, 2015, Editorial Department of Journal of Sichuan University. All right reserved.