1. Digital Library
  2. Heritage Group of Institutions, Kolkata
  3. Heritage Institute of Technology,Kolkata
  4. Publications of Academics and Others (HITK)
  5. Applied Electronics and Instrumentation Engineering (Publications )
Please use this identifier to cite or link to this item: http://localhost:80/xmlui/handle/123456789/7345
Full metadata record
DC FieldValueLanguage
dc.contributor.authorAlam Samiul, Gupta Rajarshi and Das Sharma Kaushik-
dc.date.accessioned2023-02-08T06:56:39Z-
dc.date.available2023-02-08T06:56:39Z-
dc.date.issued2023-01-24-
dc.identifier.urihttp://172.16.0.4:8085/heritage/handle/123456789/7345-
dc.language.isoenen_US
dc.publisherIEEEen_US
dc.subjectOn-Device Multi-level Signal Quality Aware Compression for Energy-efficient Wearable PPG Sensingen_US
dc.subject2023en_US
dc.subjectJanuaryen_US
dc.subject2023en_US
dc.subject24-01-2023en_US
dc.subjectAbstract— On-device computing in biomedical sensors have become attractive towards developing wearable health monitoring systems. The challenge is to make a compromise between the latency and complexity in a resource constrained implementation. This paper describes an on-device implementation of multi-level signal quality aware and quality controlled compression (MSQQCC) that enhances the compression factor while preserving the clinical features in a wearable photoplethysmography (PPG) sensing application. The multi-level quality assessment (QA) provides three eligible PPG qualities, viz., ‘excellent’, ‘good’ and ‘average’, based on which corresponding upper limits are set for further compression using a discrete wavelet transform, while the ‘corrupted’ segments are discarded. A pretrained multilayer perceptron neural network provides the optimal quantization level of coefficients. The residual data is separately compressed using an autoencoder. MSQQCC was evaluated with 300 mins of PPG data from 3 public data sets and 110 mins of data collected at laboratory. The end-to-end pipeline was implemented in a standalone system with ARM Cortex A53 controller, requiring 35.51 KB of memory and 1.8s latency to process 4s PPG data. The on-device QA achieved 98.45 % overall accuracy (Ac), which outperforms published works on PPG QA. The mean deviation of PPG clinical features by 5 %, with overall CR and PRD were 40.85 and 2.52, which are superior than many published works. A real-time transmission over Bluetooth shows improvement of energy efficiency by a significant factor and 34% extended battery life for wearable PPG sensor. The results are encouraging for adoption of MSQQCC in wearable biomedical health monitoring.en_US
dc.titleOn-Device Multi-level Signal Quality Aware Compression for Energy-efficient Wearable PPG Sensingen_US
dc.typeImageen_US
Appears in Collections:Applied Electronics and Instrumentation Engineering (Publications )

Files in This Item:
File Description SizeFormat 
2023 On-Device_Multi-level_Signal_Quality_Aware_Compression_for_Energy-efficient_Wear.pdf166.15 kBAdobe PDFView/Open
Show simple item record


Items in DSpace are protected by copyright, with all rights reserved, unless otherwise indicated.