预览加载中,请您耐心等待几秒...
1/3
2/3
3/3

在线预览结束,喜欢就下载吧,查找使用更方便

如果您无法下载资料,请参考说明:

1、部分资料下载需要金币,请确保您的账户上有足够的金币

2、已购买过的文档,再次下载不重复扣费

3、资料包下载后请先用软件解压,在使用对应软件打开

基于Context模型和矢量—标量量化器的ECG信号压缩 Abstract: ECG(Electrocardiogram)signalcompressionreferstotheprocessofreducingthesizeoftheECGsignalwhileretainingtheimportantinformation.Inthispaper,weproposeanovelmethodtocompressECGsignalsbasedontheContextmodelandVector-ScalarQuantizer.TheproposedmethodtakesintoaccounttheuniquepropertiesoftheECGsignalsandachieveshighcompressionratioswithminimallossofinformation.TheexperimentalresultsontheMIT-BIHarrhythmiadatabaseshowthattheproposedmethodoutperformsexistingcompressionmethodsintermsofcompressionratioandsignalquality. Introduction: ECGsignalsarewidelyusedinthediagnosisofheart-relateddiseases.TheECGsignalisatime-varyingsignalthatrepresentstheelectricalactivityoftheheartovertime.TheECGsignalcontainsimportantinformationrelatedtothecardiacrhythmandmorphology.DuetothehugeamountofECGsignalsgeneratedeachday,thestorageandtransmissionofECGsignalsbecomeachallengingtask.Therefore,ECGsignalcompressionisessentialtoreducethestoragespaceandtransmissionbandwidth. NumerousmethodshavebeenproposedintheliteratureforECGsignalcompression.However,mostofthesemethodsdonottakeintoaccounttheuniquepropertiesoftheECGsignals,suchasthenon-stationary,non-Gaussian,andnon-linearproperties.Therefore,thereisaneedforacompressionmethodthatcancapturetheuniquepropertiesoftheECGsignalsandachievehighcompressionratioswhilepreservingtheimportantfeaturesofthesignal. ContextmodelandVector-ScalarQuantizer(VSQ)aretwowidelyusedtechniquesforsignalcompression.Thecontextmodelisastatisticalmodelthatcapturesthecorrelationbetweentheneighboringsamplesinthesignal.TheVSQisaquantizationtechniquethatcanachievehighcompressionratioswhileminimizingthedistortion. Inthispaper,weproposeanovelmethodforcompressingECGsignalsbasedonthecontextmodelandVSQ.TheproposedmethodtakesintoaccounttheuniquepropertiesoftheECGsignalsandachieveshighcompressionratioswithminimallossofinformation. Methodology: Theproposedmethodconsistsoftwomainstages:trainingandcompression.Inthetrainingstage,thecontextmodelistrainedtocapturet