ECG Signal Recovery Problem Based on Compressed Sensing Theory

Body sensor network (BSN) is a new network which can monitor the patient’s physiological characteristics in real time, thus the network is widely utilized to improve the level of healthcare for people through the computer techniques, such as intelligent information processing, new network services and so on. Compressed sensing (CS) is a new theory of signal acquisition and decoding, which not only breaks through the bottleneck of the law of traditional sampling, but also makes the signal sampling and compression done at the same time. So, CS is extensively used in medicine, astronomy, pattern recognition and so on. For now, CS is paid more and more attention to by many researchers. This paper faces the recovery problem of electrocardiogram (ECG) signals through compressed sensing theory. According to the various performance of all kinds of wavelet basis conducted as sparse basis, we hope to find an optimal wavelet basis that makes the ECG signal restored precisely and stably through a lot of experiments. The experimental results demonstrate that there exists a set of wavelet basis which can make a variety of ECG signals recuperated with high accuracy and strong robustness. The result can provide ideas for discussing the recovery problem of ECG signals furthermore.