Dictionary selection for compressed sensing of EEG signals using sparse binary matrix and spatiotemporal sparse Bayesian learning

Energy efficiency is still the obstacle for long-term real-time wireless ECG monitoring. In this paper, a digital compressed sensing- (CS-) based single-spot Bluetooth ECG node is proposed to deal with the challenge in wireless ECG application. A periodic sleep/wake-up scheme and a CS-based compression algorithm are implemented in a node, which consists of ultra-low-power analog front-end, microcontroller, Bluetooth 4.0 communication module, and so forth. The efficiency improvement and the node’s specifics are evidenced by the experiments using the ECG signals sampled by the proposed node under daily activities of lay, sit, stand, walk, and run. Under using sparse binary matrix (SBM), block sparse Bayesian learning (BSBL) method, and discrete cosine transform (DCT) basis, all ECG signals were essentially undistorted recovered with root-mean-square differences (PRDs) which are less than 6%. The proposed sleep/wake-up scheme and data compression can reduce the airtime over energy-hungry wireless links, the energy consumption of proposed node is 6.53 mJ, and the energy consumption of radio decreases 77.37%. Moreover, the energy consumption increase caused by CS code execution is negligible, which is 1.3% of the total energy consumption.

Download Full-text

DCS-based MBSBL joint reconstruction of multi-sensors data for energy-efficient telemonitoring of human activity

International Journal of Distributed Sensor Networks ◽

10.1177/1550147718767612 ◽

2018 ◽

Vol 14 (3) ◽

pp. 155014771876761

Author(s):

Jianning Wu ◽

Jiajing Wang ◽

Yun Ling

Keyword(s):

Energy Efficiency ◽

Compressed Sensing ◽

Sparse Representation ◽

Human Activity ◽

Energy Efficient ◽

Bayesian Learning ◽

Sparse Bayesian Learning ◽

Distributed Compressed Sensing ◽

Joint Reconstruction ◽

Sparse Representation Classification

The joint reconstruction of nonsparse multi-sensors data with high quality is a challenging issue in human activity telemonitoring. In this study, we proposed a novel joint reconstruction algorithm combining distributed compressed sensing with multiple block sparse Bayesian learning. Its basic idea is that based on the joint sparsity model, the distributed compressed sensing technique is first applied to simultaneously compress the multi-sensors data for gaining the high-correlation information regarding activity as well as the energy efficiency of sensors, and then, the multiple block sparse Bayesian learning technique is employed to jointly recover nonsparse multi-sensors data with high fidelity by exploiting the joint block sparsity. The multi-sensors acceleration data from an open wearable action recognition database are selected to assess the practicality of our proposed technique. The sparse representation classification model is used to classify activity patterns using the jointly reconstructed data in order to further examine the effectiveness of our proposed method. The results showed that when compression rates are selected properly, our proposed technique can gain the best joint reconstruction performance as well as energy efficiency of sensors, which greatly contributes to the best sparse representation classification–based activity classification performance. This has a great potential for energy-efficient telemonitoring of human activity.

Download Full-text