Application Research Of Chaotic Encryption In Real-time Privacy Data

The problem of sensitive data leakage accompanying the entire life cycle of big data collection,analysis,transmission,and storage is becoming more and more serious.Cryptography is one of the effective ways to ensure data security.Chaotic cryptography is a cross-integration of chaos theory and cryptography.In the research and exploration of many scholars at home and abroad for 30 years,in image encryption and digital Some achievements have been made in the field of information security such as watermark and Hash function,but there is not much research on the combination of chaotic cryptography and privacy protection of big data.It is also one of the hot topics in current research.Spark Streaming is a stream data processing framework built on the lightweight data processing platform Spark.It can be compatible with both batch processing and real-time data processing logic and algorithms,and is widely used in various real-time data analysis processing scenarios,but the processing of streaming data does not provide a good privacy protection scheme.In order to realize the identification and protection of sensitive data in massive data streams,this paper combines chaotic cryptography and big data real-time processing framework Spark Streaming to make some attempts and explorations,and designs a real-time privacy data chaotic encryption system based on Spark Streaming(RPCES).The specific work is as follows:(1)A 5-dimensional non-degenerate high-dimensional discrete-time hyperchaotic system is designed.Firstly,a nonlinear nominal matrix is given,and then a similar transformation is obtained by non-singular matrix to obtain an asymptotically stable nominal system.Then a uniformly bounded controller and control matrix are designed.The chaotic anti-control is used to configure the poles of the system,so that the system has five large positive Lie indices and achieves no degeneracy.(2)A Spark Streaming Real-Time Privacy Data Chaotic Encryption System(RPCES)was designed.The system consists of four parts: data collection system,data analysis and screening system,data encryption system,and data storage system.Among them,in the data analysis and screening system,through the machine learning algorithm in the Spark MLlib library owned by the big data platform,the privacy data screening model is trained for data classification.In the data encryption system,data encryption is performed using a chaotic encryption algorithm.The entire system is scalable and real-time,highly fault tolerant and secure.(3)A high-dimensional discrete-time hyperchaotic encryption algorithm is proposed.The encryption algorithm combined with the Spark Streaming mechanism can encrypt the data in parallel,and the algorithm design realizes the closed-loop feedback of the encrypted ciphertext through the ring coupling,which can effectively resist the known plaintext and select the plaintext attack.The chaotic sequence generated by the algorithm can pass the extremely strict TESTU01 test.The results show that RPCES has good scalability,stability and real-time processing ability,which can effectively identify sensitive data in massive data streams and reduce the amount of data required for encryption.The use of chaotic encryption algorithm to encrypt sensitive data in real time can effectively reduce the probability of privacy data leakage in massive data streams and ensure data security.