基于ECG的活体检测与身份验证SOC设计  

SOC Design for ECG-based Liveness Detection and Biometric Human Identification

作　　者：王永蔹 陈子为[1] 曹坤 杨玉航 谢帅 WANG Yonglian;CHEN Ziwei;CAO Kun;YANG Yuhang;XIE Shuai(College of Electronic Engineering,Chengdu University of Information Technology,Chengdu 610225,China)

机构地区：[1]成都信息工程大学电子工程学院,四川成都610225

出　　处：《成都信息工程大学学报》2023年第5期543-547,共5页Journal of Chengdu University of Information Technology

基　　金：国家级大学生创新创业训练计划资助项目(S202110621086);四川省科技计划重点研发资助项目(2019YFG0126)。

摘　　要：由于心电图的活体指示特性和独特而复杂的信号特征,采用心电图进行身份识别是最安全的生物识别方法之一。提出了一种在FPGA上搭建适用于ECG身份识别的专用片上系统(SOC)设计方案。该设计利用基于主成分分析(PCA)与欧几里得距离度量的ECG身份识别算法进行身份识别,并利用FPGA并行运算的优势对该识别算法实现硬件加速。最后基于ARM公司开源的DesignStart Cortex-M3 IP核,在Xilinx FPGA上实现了该片上系统。结果显示:所设计的片上系统识别正确性可达96.8%,运行性能最高可达90 MHz,满足实时性需求。Electrocardiogram(ECG)is one of the most safety-relevant biometrics due to its live indication and the unique specific waveform.This paper proposes a complete System-on-Chip(SOC)design on FPGA for human identification using Electrocardiograms(ECG)biometric.The ECG identification algorithm based on PCA(principle component analysis)and Euclidean distance metric is used for human identification,and the hardware acceleration of the identification algorithm is realized by using FPGA parallel operation.Based on the ARM DesignStart Cortex-M3 IP core,a SOC is built on Xilinx FPGA.The achieved implementation results show that the proposed SOC reached an identification accuracy of 96.8%and operating performance up to 90 MHz,which met the realtime processing requirements.

关 键 词：DesignStart 片上系统 生物特征人体识别 ECG 

分 类 号：TP391.41[自动化与计算机技术—计算机应用技术]