Performance improvement of PAM-DS UWB signal in AWGN channel based on EMD

The empirical mode decomposition (EMD) method for analysing nonlinear and non-stationary data has attracted great research interests. Because, the PAM-DS UWB signal which has the property of transient and low probability of intercept (LPI) is non-stationary, in this paper, the EMD method is used to extract the weak impulse radio signal from noisy environment. The noisy PAM-DS UWB signals are decomposed into a finite number of oscillatory modes known as intrinsic mode functions (IMFs) and a residue. Based on different time scale characteristic of the signal and noise, a majority of noise will be removed from signal by proper selection of IMFs. An efficient elimination of noise on UWB signal and a great improvement on Bit Error Rate (BER) are achieved by simulations.