A speech enhancement algorithm based on the Empirical mode decomposition (EMD) and improved spectral subtraction is proposed for the low SNR (Signal Noise Ratio) short-wave time signal. This method is proposed to solve the problem that the shor-twave time signals cannot be used for timing in complex noise environments. The core idea of this method is to use the Hilbert-Huang Transform (HHT) algorithm to decompose the noised short-wave signals with the empirical mode decomposition, and to select the intrinsic mode functions containing the shor-twave signals reconstruction through the maximum correlation. Then the reconstructed signal is spectrally subtracted to achieve the purpose of noise reduction. Experiments show that this method has a better noise reduction than the traditional methods.