中文摘要：

基于时空稀疏模型,提出一种穿戴式心电信号的压缩感知方法,利用信号的时间相关性和空间相关性,来实现心电信号的重构。同时,还提出了一种＂分—合＂式字典学习算法,通过利用心电信号内在的聚类结构,对训练样本进行字典学习,从而构造出符合心电信号特点的字典,并对其进行稀疏表示。从而进一步提高了心电信号的重构性能。为了验证提出的心电信号压缩感知方法的有效性,采用OSET数据库中的心电数据,将其与其他两种基准算法进行了对比。仿真实验结果表明,所提出的心电信号压缩感知方法能有效地提高心电信号重构的质量。

英文摘要：

A spatio-temporal sparse model-based method is proposed for the compressive sensing of electrocardiosig- nal. The eleetroeardiosignal is reconstruted by exploiting the temporal and spatial correlation of signal. In addition, a ＂split-merge＇dictionary learning approach is developed. It determines a dictionary by using its inherent clustered structure, and the electrocardiosignal is sparse represented on this dictionary. Thus, the reconstruction performance of eleetrocardiosignal is further improved. The proposed compressive sensing method of electroeardiosignal is com- pared with other two benchmarking methods to illustrate its effectiveness. The simulation results show the proposed method can improve the quality of electrocardiosignal reconstruction.