Application of an Approximately Shift-invariant Wavelet Transform to Signal Detection via Empirical Model of Noise

This article proposes a use of the dual-tree discrete wavelet transform (DT DWT) to the problem of signal detection in underwater sound. This approximately shift invariant DT DWT can generate multi-resolution subspaces that keep more of their coefficient energy in each of these subspaces than discrete wavelet transform´s (DWT). The detection performance comparison, under the same value of the false alarm probability P_F, of this proposed DT DWT-based empirical model and DWT- based method is presented. A performance comparison has shown that this proposed DT DWT-based empirical model of noise is better than the DWT-based detector.