Hybrid Blockchain Protection Method Based on Homomorphic Encryption and Machine Learning
In order to ensure the security of user distribution,transaction and change operations in hybrid blockchain,a hybrid blockchain privacy protection method based on homomorphic encryption and machine learning is proposed.Based on the public key compression and module transformation,the homomorphic encryption method is adopted to generate a homomorphic en-cryption hybrid blockchain through key generation,encryption,ciphertext operation,decryption and inspection,so as to realize the data encryption of the hybrid blockchain.The forward propagation method of convolutional neural network based on ma-chine learning is adopted to complete the homomorphic encryption result verification through the operations of convolutional layer,pooling layer,incentive layer and global average pooling layer in turn.After passing the inspection,this paper builds a privacy hybrid blockchain to enable privacy data ciphertext to be linked,and eliminate invalid data before linking,so as to real-ize hybrid blockchain protection.Experimental results show that the hybrid blockchain data encryption method has high effi-ciency and inspection accuracy.The success rate of privacy protection is close to 100％and is not affected by the increase in the number of nodes.It can ensure the security of hybrid blockchain distribution,transaction and change operations,and effective-ly avoid the disclosure of users'privacy information.
万方数据
维普
