Research And Implementation Of Identification Algorithm Based On ECG Information

Nowadays science and technology are developing more and more rapidly, the traditional identification methods are facing a huge challenge made by emerging technology. Some ordinary identity authentication techniques such as static password could be cracked easily. The electrocardiogram(ECG) signal that this article studies is one of the inherent physiological characteristics of human body. The mechanism of ECG formation is very complex, so it is difficult to be counterfeited. Moreover, the signal composition varies from person to person. Therefore, it is safe and reliable that ECG information is used in the research field of identity recognition.The wearable ECG and photoplethysmography(PPG) synchronous acquisition devices are used to collect data. This paper exactly locates a peak of R wave which is the most distinctive characteristic of ECG signal on the basis of relative positions of PPG and ECG signal. In order to extract the peak of R wave in the most accurate R wave invert detection and interference judgment methods are designed. On the basis of the hybrid of ECG signal transformational features and analytical features, the paper designs a twice-matching method for identification which improves the recognition rate greatly. And then, the software of ECG identification is also designed in this paper on android platform, the functions contain bluetooth connectivity, user registration, user identification and so on.Besides the data collected in our laboratory, data from the public databases are also used for verification of algorithm described in this paper. Identification rate is calculated based on the testing set as the time for 6s and 30 s respectively. The experimental results show that the algorithm is adaptable, as well as the correct identification process implemented by android application designed in this paper. In all, this article combines theory with practice closely.