The Application Of ECG Identification Technology For Car Security System

The biological feature information with uniqueness and safety has become a new medium for identity authentication. The ECG signal is a kind of human biometric information and it is focused in this research. Since the 21 st century, some researchers begin to explore the ECG identification technology, and had a few achievements, but the actual application has not spread yet. This paper discusses how to put the ECG identification technology on the automotive safety system properly, in order to solve the security problems of the automobile anti-theft issue.This paper designs a suitable ECG identification system to be embedded in the vehicle systems. Within that, the hardware circuit design and software algorithm research work are performed. The hardware circuit design includes the design of the acquisition module and the main controller. In the design of the acquisition module, the amplifying circuit and the filter circuit part are mainly discussed. In the design of the main controller, the computing speed, the storage space, and the interface requirements are mainly considered. The research on software algorithms mainly include the recognition algorithm,the denoising algorithm of ECG signal, and the recognition problem after exercise.In the ECG denoising, this paper designs an experiment to compare the FFT denoising and the wavelet denoising, the result finds out that the wavelet denoising is better. The identification algorithm is mainly carried out in the frequency domain. The ECG recognition problem of motion state, is mainly discussed from the two directions of the model prediction and multi sample database. The main achievements of this paper are listed below:(1) Design a suitable embedded ECG identification for vehicle system, and then implement its hardware circuit and software algorithm.(2) According to the characteristics of weak and noise signal, we design the amplifier circuit and filter circuit, and the de-noising processing software properly, the good results are achieved.(3) Proposes a recognition algorithm based on the frequency domain and conducts its experimental design as well. In order to solve the problem of the uncomplete waveform, we analyze spectrum characteristics in seeking for several cycles before the average waveform.(4) Discovers that the establishment of multi-rhythm sample database, it can improve the accuracy of ECG recognition under the motion state.The design of the system with CAN bus interface, can be integrated well with the automotive control system. This can make the identification system control the car conveniently after correct recognition. On the future development of this system, we can put the disease diagnosis technology of ECG based on fusion method. It can monitor the driver's heart disease, and has a significance to the safety of driving a car.