Research on target signal detection based on neural networks and wavelet decomposition

This paper deals with feature extraction in combination of wavelet decomposition with neural networks, under the condition of low SNR. The signal is first decomposed by wavelet transform, and on the base, the decomposed coefficients are reconstructed to form a new time series, from which some energy parameters can be extracted by time-domain analysis. By means of BP network, it is possible to recognize whether target signal is involved or not in received signals. The effectiveness of the method is verified by a real target signal with additive simulated noise signal, especially under the condition of low SNR.