A single-channel BSS new method of time-frequency overlapped communication signal based on multi-scale CWT and ICA

To solve the single-channel blind separation problem of time-frequency overlapped communications signals, this paper proposes a new method joint multi-scale continuous wavelet transform (CWT) and independent component analysis (ICA). By using multi-scale CWT decomposition on single-channel signal, the original mixed signal can be expanded to a multi-dimensional signal without any prior knowledge, then use the ICA algorithm to realize its blind separation. Next, determine the number of source signals and eliminate the redundancy and irrelevancy to complete the separation. Finally, the separation experiments and simulations confirmed this proposed method valid and feasible.