基于块稀疏贝叶斯学习的体域网心电压缩采样

发布时间：2017-12-27 06:36

本文关键词：基于块稀疏贝叶斯学习的体域网心电压缩采样　出处：《传感技术学报》2015年03期 　论文类型：期刊论文

【摘要】：为有效提高体域网的实时性和降低体域网的功耗,提出一种基于块稀疏贝叶斯学习的体域网心电压缩采样方法。该方法在体域网框架下,利用压缩采样理论,在体域网的传感节点利用二进制随机观测矩阵对心电信号进行压缩采样,远程监护中心获得采样值之后,利用块稀疏贝叶斯学习重构算法和离散余弦稀疏变换矩阵对心电信号进行重构。实验结果表明,当心电信号压缩率在70%~90%时,基于块稀疏贝叶斯学习的重构算法要比其他重构算法的重构信噪比高出3 d B~21 d B。该方法能有效减少数据采样,减轻后续的数据存储、数据传输压力,提高体域网的实时性。同时该方法具有功耗低,易于硬件实现的优点。
[Abstract]:In order to effectively improve the real-time performance of the body area network and reduce the power consumption of the body area network, a somatic area network ECG compression sampling method based on block sparse Bayesian learning is proposed. The method in the body area network framework, the use of compressed sampling theory, the sensor node network using a binary random measurement matrix for ECG compression after sampling, remote monitoring center for sampling, using block sparse reconstruction algorithm and discrete cosine Bias learning sparse transform matrix to reconstruct the ECG signal. The experimental results show that when the compression rate of electrical signals is 70%~90%, the reconstruction algorithm based on block sparse Bayesian learning is 3 D B~21 D B higher than that of other reconstruction algorithms. This method can effectively reduce the data sampling, reduce the subsequent data storage, data transmission pressure, and improve the real time of the body area network. At the same time, the method has the advantages of low power consumption and easy hardware implementation.
【作者单位】：南昌大学机器人研究所;江西财经大学软件与通信工程学院;
【基金】：国家自然科学基金项目(61273282) 江西省高等学校科技落地计划项目(KJLD13002)
【分类号】：TN911.7
【正文快照】： 人口老龄化的加剧、“空巢”家庭数量的增加以及大量伤残病人的出现给社会带来了诸多问题,其中一个重要问题就是如何为老人及病人提供实时有效的远程健康监护服务。体域网BSN(Body Sensor Net-work)是解决该问题的较好方案[1-3]。心电ECG(E-lectrocardiogram)信号,作为人体最

【共引文献】