WANG Yonglian,CHEN Ziwei,CAO Kun,et al.SOC Design for ECG-based Liveness Detection and Biometric Human Identification[J].Journal of Chengdu University of Information Technology,2023,38(05):543-547.[doi:10.16836/j.cnki.jcuit.2023.05.008]
基于ECG的活体检测与身份验证SOC设计
- Title:
- SOC Design for ECG-based Liveness Detection and Biometric Human Identification
- 文章编号:
- 2096-1618(2023)05-0543-05
- 关键词:
- DesignStart; 片上系统; 生物特征人体识别; ECG
- Keywords:
- DesignStart; system on chip; biometric human identification; ECG
- 分类号:
- TP391.41
- 文献标志码:
- A
- 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。
- Abstract:
- 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.
参考文献/References:
[1] Zhai X,Amira A,Bensaali F.ECG Security IdentificationSystem on the Zynq SoC Platform[C].IEEE InternationalConference on Computer & Information Technology; Ubiquitous Computing & Communications Dependable.IEEE,2015:1164-1169.
[2] A.George,S.Marcel.Learning One Class Representations for Face Presentation Attack Detection Using Multi-Channel Convolutional Neural Networks[J].IEEE Transactions on Information Forensicsand Security,2021,16:361-375.
[3] Melzi P, Tolosana R, Vera-Rodriguez R. ECG Biometric Recognition: Review, System Proposal, and Benchmark Evaluation[EB/OL].
[4] David Pereira Coutinho,Hugo Silva,Hugo Gamboa, et al. Novel fiducial and non-fiducial approaches to electrocardiogram-based biometric systems[J].IET Biom.,2013,2:64-75.
[5] 郭炜,魏继增,郭筝,等.SoC设计方法与实现[M].2版.北京:电子工业出版社,2011.
[6] 杨翠娥.基于FPGA的片上系统设计研究[J].山西电子技术,2021(4):82-84.
[7] Uwaechia A N,Ramli D A.A Comprehensive Survey on ECG Signals as New Biometric Modality for Human Authentication:Recent Advances and FutureChallenges[J]. IEEE Access,2021,9:97760-97802.
[8] Zhang Q,Zhou D,Zeng X.HeartID:A Multiresolution Convolutional Neural Network for ECG-Based Biometric Human Identification in Smart Health Applications[J].IEEE Access,2017,5:1805-11816.
[9] A.Ait Si Ali,X.Zhai,N.Ramzan,et al.Enhanced Biometric Security and Privacy Using ECG on the Zynq SoC[J].Biometric Security and Privacy,Signal Processing for Security Technologies,2017,4:179-202.
[10] 陈龙,陈子为,朱美吉,等.基于FPGA的实时图像去雾系统[J].成都信息工程大学学报,2021,36(2):138-142.
[11] 陈艺.一种基于FPGA软核处理器的数字控制系统设计[J].电子世界,2019(14):165-166.
[12] ARM Inc.ARM Cortex-M3 Designstart Eval RTL and Testbench UserGuide [EB/OL].
[13] ARM Inc.Cortex-M System Design Kit Technical Reference Manual [EB/OL].
[14] Alex Forencich.Verilog AXI Components[EB/OL].
[15] Xilinx Inc.(2012).Vivado HLS User Guide[EB/OL].[accessed on 30-March-2022].
备注/Memo
收稿日期:2022-09-13
基金项目: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)