Compressed sensing for multi-lead electrocardiogram signals

Compressed sensing is widely used due to its ability to reconstruct the signal accurately from a set of samples which is smaller than the set of samples produced using Nyquist rate. Multi-lead electrocardiogram signals show sparseness in wavelet domain. In this work, compressive sensing is applied for electrocardiogram signals in transform domain using random sensing matrix with independent identically distributed (i.i.d.) entries formed by sampling a Gaussian distribution. The reconstruction of sparsely represented signal is performed by convex optimization problem by L1-norm minimization. The quality of processed signal is satisfactory. Signal distortions are evaluated using percentage root mean square difference (PRD), root mean square error (RMSE), normalized root mean square difference (NRMSD), normalized maximum amplitude error (NMAX) and maximum absolute error (MAE). The lowest PRD value, 1.723%, is found for lead-V5 signal at sparsity level of 26.76%, using database of CSE multi-lead measurement library for simulation.