A method for specific emitter identification based on empirical mode decomposition

Real radio and radar signals turn out to be non-stationary and nonlinear time series, and estimating instantaneous parameters of such signals in time domain will benefit detecting and identifying specific emitters, such as civil or military radios. However, conventional methods such as Wavelet and Wigner-Ville distribution (WVD) methods are difficult to deal with these transient signals when the signal is no-stationary. To provide more efficient approach for estimating instantaneous parameters of non-stationary and identifying specific emitter with high accuracy, a new method based on empirical mode decomposition (EMD) for analyze no-stationary signal is proposed in this paper. The results show that the instantaneous frequency estimates generated by the EMD method are more accurate than those obtained by the Wavelet technique.