Interference Detection And Quality Assessment Of Pulse Signal On Android Platform

The collection process of the pulse signal by mobile intelligent monitoring platform can be easily interfered by many factors,which leads to the low quality of the acquired pulse signal and affects its results on the health monitoring system.In order to solve this problem,the corresponding pulse signal interference detection algorithm based on the causes of interference and the manifested characteristics of interference in pulse signal acquisition was proposed.Based on this algorithm,a Cascaded Layer-by-Layer Discrimination method is proposed to evaluate the quality of pulse signal,by which pulse signal is divided into available signal and unavailable signal.Experimental results on PC and Android platform show that the proposed algorithm can detect the interference segments accurately in the pulse signal in real time,and reliably evaluate the availability of the pulse signal.In addition,the proposed method is easy to implement and popularize,which provides reliable technical support for the health monitoring system based on mobile intelligent terminal.The main research findings of the content are as follows:(1)Pulse signal acquisition and transmission platform construction.A pulse sensor located on the finger was developed to achieve accurate and stable pulse signal acquisition.The human pulse signal is collected by a sensor,and then I/V conversion,filtering,amplification and other operations were carried out by a pulse signal conditioning circuit;and finally,the processed signal was converted into a digital signal through A/D of a single chip microcomputer,and then was transmitted via Bluetooth to the handset for signal processing and analyzing.(2)The pulse signal filtering.The pulse signal is kind of a weak signal,and it is easy to be mixed some interference signals during the acquisition process.such as frequency interference,electromyography interference,baseline drift noise interference signal,these signals show certain characteristics in frequency,so can be suppressed or filtered by filtering method.In order to restrain or eliminate the interference and noise,this paper analyzed the causes and manifestations of pulse signal noise interference was analyzed.Then comprehensively considered the real-time performance of Android platform,used integer coefficient filtering method to filter out these noise interference signals.(3)The pulse signal interference detection and quality evaluation.Some of those interfere signals present in the pulse signal have no specific frequency characteristics,such as pulse interference generated during sensor slip and placement,pulse signal lost segments,and motion artifacts generated by the sensor as it slides over the body surface,these cannot be eliminated or identified by filtering.A pulse signal interference segment detection algorithm was proposed according to the characteristics of interference waveform,and was compared with the traditional DTW interference segment detection algorithm and SVM-based interference segment detection algorithm.Experimental results show that the proposed method can meet the requirements of Android client in accuracy and real-time.Finally,on the basis of interference detection,a cascade discrimination method of pulse quality evaluation was proposed,which separated the pulse signal into available signal and unavailable signal,so as to remove the unavailable signal and reserve the available signal.(4)Android phone client program design.A mobile phone software platform was designed to verify the usefulness of the proposed algorithm.The mobile phone terminal was connected with the pulse acquisition device through bluetooth,and carried out the operation of signal processing,data storage and waveform display on the acquired pulse signal,and emphatically verified the proposed method.The experimental results show that the proposed method is easy to implement and versatile.