Research On Key Generation Based On ECG Signal

With the development of science and technology,biological characteristics are playing an increasingly important role in people's lives.Because of its uniqueness,peculiarity,and randomness,it is widely used in digital certificates,information encryption,and key protection.Biometric cryptography,that is,the combination of the biometric information of users and the secret information that needs to be kept safely,is to protect the secret information and avoid the disclosure of biometric information of users.The generation mechanism of ECG signals is complicated and difficult to be imitated and forged,and its security is high.Therefore,this paper proposes a method to combine the characteristic information of ECG signals with cryptography and mathematics to achieve key generation and the purpose of key binding.The main problem of using ECG signals as keys is to find the stable characteristics of ECG signals.In view of this problem,this paper proposes a method of superposing individual single cardiac cycles to form a superposition matrix.That is,by identifying the R point of each single cardiac cycle of an individual,the waveform of the individual ECG signal is divided into individual single cardiac cycles,and then multiple single cardiac cycles are superimposed to form a superposition matrix.This method can increase the weight of the aggregation points in the matrix,thereby improving the division of the stable region of the ECG signal.The superimposed matrix can be used to extract high-level location area information in the matrix-mapped 3D graphics,and record its location information to form a set of stable features.Experiments show that with the increase of the number of superimposed bars,the stable area will tend to a fixed position,and the individual stability characteristics at this fixed position will become more obvious.In order to solve the problem of how to use the key after the key is generated,this paper proposes a fuzzy vault scheme based on ECG signals.This scheme combines biometrics and cryptography and uses reconstruction polynomials to perform operations.The method for finding the stable characteristics of ECG signals and the secret information that needs to be encrypted are combined with cryptographic knowledge to form a fuzzy vault.In the unlocking(decryption)process,it is only necessary to compare the same points in all the feature information.Experimental results show that when the dimension of the stable point sequence selected by the individual is 30,and the number of reconstruction polynomials selected in the fuzzy vault is 4,5,6,and 7,the average unlock(decryption)pass rates are 82.86%,77.14%,75% and 70%,When the dimension of the stable point sequence selected by the individual is 50,and the number of reconstruction polynomials selected in the fuzzy vault is 4,14,18,24,the average unlock(decrypt)pass rates are 95.71%,90%,82.14% and 61.43%.It can be seen that the unlocking rate is directly proportional to the dimension of the selected stable sequence,and inversely proportional to the degree of the reconstruction polynomial.In practical applications,the dimension of the stable point sequence and the degree of the polynomial must be selected according to the number of bits required to generate the key,so as to ensure the balance of unlocking rate and security.