Adaptive encryption method of sensitive data in data center database based on big data cross-mapping fusion algorithm

Traditional methods are inadequate for addressing the requirements of data security and efficiency in contemporary data centers. To address the problems of security and transmission efficiency of sensitive data in data center databases, this study proposes an adaptive encryption method for sensitive data in data center databases based on big data cross-mapping fusion algorithm. Sensitive data collection in data center databases is achieved through sparse representation, compressed measurement, and the use of compressed sensing to recover and reconstruct data. A self-organizing mapping neural network (SOM) is used to perform sensitive data fusion through four steps: Matching, clustering, weight updating, iterative competition, and mapping. The process is optimized by employing the cross validation (CV) method to enhance the efficacy of sensitive data fusion.The improved AES algorithm is used to establish the optimal affine transformation for generating a novel S-box, and is integrated with the squared residual algorithm to implement the key extension of the AES algorithm. This approach transforms the serial operation structure of the AES algorithm into a parallel operation structure, enabling adaptive encryption of fusion sensitive data. Experimental results demonstrate that the security of sensitive data can reach 99% after encryption using this method. This approach facilitates the encryption of sensitive data within the data center and enhances the security of sensitive data in the data center