Hardware Implementation of Non-Linear Quantization Algorithm for a Wavelet-based ECG Data Compress System

碩士 === 國立高雄第一科技大學 === 電腦與通訊工程所 === 95 === Electrocardiogram (ECG) signal analysis is a non-invasive modality widely used for heart disease clinical diagnosis .The diagnosis of cardiac patients depends on the quality of the recording signal or effective real-time data transmission. ECG compression sy...

Full description

Bibliographic Details
Main Authors: qiao-wen lie, 劉喬溫
Other Authors: King-Chu Hung
Format: Others
Language:zh-TW
Published: 2007
Online Access:http://ndltd.ncl.edu.tw/handle/47141846326653021820
Description
Summary:碩士 === 國立高雄第一科技大學 === 電腦與通訊工程所 === 95 === Electrocardiogram (ECG) signal analysis is a non-invasive modality widely used for heart disease clinical diagnosis .The diagnosis of cardiac patients depends on the quality of the recording signal or effective real-time data transmission. ECG compression system and auto-recording signal is good choices. Associating the RROLT with the 1-D NRDPWT,1-D RRO-NRDPWT apply to ECG compression systems successfully. This non-linear quantization algorithm base on 1-D RRO-NRDPWT can obtain a superior compression performance. The algorithm have two character. First, by adjust the single quantization factor (QF). Section, this algorithm impact factor (QF) attempts to keep PRD and CR in linear relationship. And non-linear quantization algorithm make a great impact on ECG compression systems other than improving compression performance of whole system. Besides,it is based on quantization error of frequency-domain aspiration to base on quality prediction of time-domain and converge on linear error predetermination model,fast. We formulated an adaptive and real-time compression mechanism ECG compression systems architecture base on a linear error predetermination model and dynamic quantization algorithm. Quantization process of this architecture is use of integer multiplication. Component of multiplication make a great impact on quantization process performance, crucially. We proposed a novel recursive multiplier. It has error of linear and low power performance to content with portable application. The MIT-BIH arrhythmia database is used for testing signals and experimental results show that quantization scale of finite word-length between academic value can obtain CR-PRD curve to save for error of bias, only. The novel recursive multiplier can obtain a superior hardware performance over traditional recursive multiplier and the area with power consumption is improved by 39% with 32% ,respectively, but the critical path delay 3%.However, it can achieve aim of the portable quality of the recording signal.