Performance evaluation of compressive sensing based compression of multi-resolution ppg signals under wban environment

Abstract

Wireless Body Area Network (WBAN) is a collection of wireless biosensors worn on the body, in which each sensor node is capable of computing and communicating with other nodes or devices like smart phones, Personal Digital Assistant (PDA), hand held devices etc,. The wearable nodes are powered by battery and need to be always functional for continuous remote monitoring of patients which demands that node life time has to be prolonged to the maximum extent. One of the best solutions for this issue is to go for data compression at the node. In this context, Compressive sensing (CS) based energy efficient compression algorithms have been developed and tested for 8-bit and 10-bit resolution Photoplethesmogram (PPG) signal. Test data have been acquired from normal subjects using Arduino Uno R3. Validation of the algorithm has been carried out by applying on MIMIC-II database and acquired signals. It is found that the CS algorithm reconstruction quality diminishes for low resolution PPG data.