| The strengthening of information security becomes more and more important with the development of information technology today.As one of the information security technology,identification has been widely permeated into all areas of our lives.However,conventional identification methods based on passwords,documents and keys are easy to be copied,stolen or forgotten,which can not meet people's needs for security.Thus,biometric identification technology appears and develops gradually.Nowadays,the biometric technologies based on fingerprint,face,iris and other characteristics develop rapidly and have been successfully applied in commerce.However,these technologies still remain some shortcomings,so it is urgent to find new biological features to replace the existing biometric technologies.In this dissertation,the biometric identification based on PhotoPlethysmoGraphy(PPG)is studied,which mainly includes the following two parts:(1)A PPG biometrics system based on NRF51822 is built.Reflective photoelectric switch is used as sensor.Firstly,the original signal is preprocessed by amplifying circuit,50HZ trap circuit and low-pass filter circuit.Then the processed signal is sent to NRF51822 for AD conversion.Finally,the signal is sent to computer for identification by Bluetooth.(2)The biometrics technology based on PPG and its influence factors are discussed.Firstly,feature points are detected by extreme value method and filtered by the stepwise selection method.Secondly,Support Vector Machine(SVM)and K Nearest Neighbor(KNN)classifier are studied for authentication and identification based on PPG feature.Then,the influences of time variability of PPG and individual motion on PPG recognition accuracy are studied.The experimental results show that the recognition rate is less affected by the time variability of PPG signal in the calm state,but largely influenced by the individual motion.Besides,the recognition rate of authentication based on PPG can reach more than 92.5%. |
PDF Full Text Request