Compressive sensing based ECG telemonitoring with personalized dictionary basis

Compressive sensing is a novel signal processing technique that addresses the energy and telemetry constraints in Wireless Body Area Network. However, common analytical basis cannot sparsify the electrocardiography signal well, and causes performance degradation in compressive sensing reconstruction. In this paper, we apply dictionary learning to construct the personalized basis for compressive sensing reconstruction. The results show that the proposed personalized basis improves the compression ratio by 2.11x compared with existing works. Moreover, considering the change of signal characteristic, we propose the physiological variation detection technique to maintain high compression ratio.