Optimized compression and recovery of electrocardiographic signal for IoT platform

Fei Yun Wu, Kunde Yang, Xueli Sheng

Research output: Contribution to journalArticlepeer-review

12 Scopus citations

Abstract

Internet of Things (IoT) promises to a continuous, remote recording and monitoring of electrocardiogram (ECG). Thus it creates large volumes of data for healthcare purposes. The huge recordings result in the heavy burden of the communication, and the wearable devices require frequent charging since the huge data consumes energy quickly. To ameliorate this, we aim to compress the recordings and, in turn, to boost the battery life. We propose a new framework from two aspects: First, an optimization solution is proposed for the measurement matrix, which uses the shrinking singular value decomposition (SSVD) strategy at the compression terminal. Second, an accelerated method based on the non-uniform norm (ANN) is proposed to estimate and reconstruct the received signal. The proposed framework of the measurement matrix optimization and ANN estimator is firstly used for the monitoring of vital parameters such as electrocardiography (ECG). Experiments are conducted to confirm the superiority of the proposed SSVD and ANN methods.

Original languageEnglish
Article number106659
JournalApplied Soft Computing
Volume96
DOIs
StatePublished - Nov 2020

Keywords

  • Compressed sensing (CS)
  • Internet of Things (IoT)
  • Non-uniform norm (NN)
  • Singular value decomposition (SVD)

Fingerprint

Dive into the research topics of 'Optimized compression and recovery of electrocardiographic signal for IoT platform'. Together they form a unique fingerprint.

Cite this