The multi-parameter physiologic signal reconstruction by means of wavelet singularity detection and signal correlation

A novel approach for reconstructing lost data from correlative signals among multi-parameter physiologic signals is proposed in this paper. The approach extracts a sample form a target signal that has data lost, and then lays the sample one by one according to singularity of a reference signal that has tight correlation with the target signal, to form a reconstructed signal in which a substitution of lost data is included. Our experiments confirm that this approach is effective for reconstructing signal having typical waveform sample and random time interval variations, especially for the data reconstruction application that requires real time processing.