Generalized Canonical Discriminant Congruential Signcryption for Secure Communication with Big Data

Abstract

Big data is a collection of huge data employed to examine and extract information from large datasets. With the generation of a large volume of data, it faces severe security risks and challenges such as data leakage, malicious use, etc. Many researchers carried out their research for performing secured data communication.  But, the data confidentiality level was not improved by using existing cryptographic methods. Therefore, a generalized canonical linear discriminant-based Multiplicative congruential signcryption (GCLD-MCS) technique is introduced for secured data communication with higher data confidentiality and lesser communication overhead.

The GCLD-MCS technique performs two processes, namely data classification and secured communication. Initially, a number of data are collected from the big dataset. After that, generalized canonical statistic distributive linear discriminant analysis is carried out to examine the linear combination of data and to classify into a number of classes. After data classification, secured data transmission is performed using Multiplicative congruential Rabin cryptographic signcryption technique. Multiplicative congruential Rabin cryptographic signcryption includes three processes namely Multiplicative congruential Key Generation, Signcryption and Unsigncryption. Experimental evaluation of the proposed GCLD-MCS technique is carried out with respect to classification accuracy, data confidentiality level, data integrity, and communication overhead, with a different number of data. The   discussed results indicates that the performance of GCLD-MCS technique increases communication security with higher data confidentiality rate, integrity, and minimum overhead   than the other state-of-the-art methods.