Low-complexity compressed sensing with variable orthogonal multi-matching pursuit and partially known support for ECG signals
Journal
Proceedings - IEEE International Symposium on Circuits and Systems
Date Issued
2015
Author(s)
Cheng, Yih-Chun
Abstract
In this paper, we present low-complexity compressed sensing (CS) techniques for monitoring electrocardiogram (ECG) signals in wireless body sensor network (WBSN). We first exploit ECG properties in the wavelet domain to extend the partially known support set (PKS) so as to reduce the support augmentation and estimation efforts in the iterative recovery algorithm. Then, variable orthogonal multi-matching pursuit (vOMMP) algorithm is proposed, which uses orthogonal matching pursuit (OMP) algorithm in the first phase to effectively augment the support set with reliable supports and adopts the orthogonal multi-matching pursuit (OMMP) in the second phase to rescue the missing supports. The reconstruction performance is thus enhanced. Furthermore, the computation-intensive pseudo-inverse operation for signal reconstruction is simplified by the matrix-inversion-free technique based on QR decomposition. The performance and complexity comparisons manifest the advantages of our proposed techniques. © 2015 IEEE.
Event(s)
IEEE International Symposium on Circuits and Systems, ISCAS 2015
Subjects
Compressed Sensing (CS)
digital wavelet transform (DWT)
Electrocardiogram (ECG)
orthogonal matching pursuit (OMP)
orthogonal multi-matching pursuit (OMMP)
Description
Lisbon, 24 May 2015 through 27 May 2015
Type
conference paper