Compressive sensing of ECG signals using plug-and-play regularization

Compressive Sensing (CS) has recently attracted attention for ECG data compression. In CS, an signal is projected onto a small set of random vectors. Recovering the original from such compressed measurements remains challenging problem. Traditional recovery methods are based on solving regularized minimization problem, where sparsity-promoting prior used. this paper, we propose alternative iterative algorithm Plug-and-Play (PnP) method, which become popular imaging problems. PnP, powerful denoiser used to implicitly perform regularization, instead using hand-crafted regularizers; been found be more successful than traditional methods. work, use PnP version Proximal Gradient Descent (PGD) recovery. To ensure mathematical convergence algorithm, in question needs satisfy some technical conditions. We high-quality fulfilling condition by learning Bayesian small-sized patches. This guarantees that proposed converges fixed point irrespective initialization. Importantly, through extensive experiments, show reconstruction quality method superior state-of-the-art