基于压缩感知的脉搏信号重构方法研究

Research on PPG Signal Reconstruction Based on Compressed Sensing

为了提高动态脉搏信号检测过程中信号存储和传输的效率,减少信号中的冗余,该文结合脉搏信号的稀疏性,提出一种改进的自适应匹配追踪算法。该算法在稀疏自适应匹配追踪算法的基础上,采用变步长和双阈值判别条件用于提高估计信号稀疏度的准确性。将所提出的算法用于建模的脉搏信号和实际采集的脉搏信号,结果表明:该算法能够快速、准确地估计信号稀疏度,具有良好的抗噪性。与现有的稀疏自适应匹配追踪算法和正交匹配追踪算法相比,该算法重构速度快、精度高。

In order to improve the storage and transmission efficiency of dynamic photoplethysmography（PPG） signals in the detection process and reduce the redundancy of signals, the modified adaptive matching pursuit（MAMP） algorithm was proposed according to the sparsity of the PPG signal. The proposed algorithm which is based on reconstruction method of sparse adaptive matching pursuit（SAMP）, could improve the accuracy of the sparsity estimation of signals by using both variable step size and the double threshold conditions. After experiments on the simulated and the actual PPG signals, the results show that the modified algorithm could estimate the sparsity of signals accurately and quickly, and had good anti-noise performance. Contrasting with SAMP and orthogonal matching pursuit（OMP）, the reconstruction speed of the algorithm was faster and the accuracy was high.