Bernstein Vazirani deep neural perceptron based certificateless signcryption for secured data communication in cloud

Abstract

Present day evolutions in providing genuine security in the Cloud Computing Network havebeen made Certificateless Signcryption model exceedingly acceptable for several applications. However, security and privacy are the paramount area of interests in safeguarding data or data packet as they are said to be highly sensitive.Moreover, with conventional Certificateless Signcryption process found to be computationally laborious and with the discrete operation minimizing the advantages gained from smaller key size, data security is said to be compromised. To address on this issue, in this work, a novel method called, Bernstein Vazirani Deep Neural Perceptron-based Certificateless Signcryption (BVDNP-CS) for secured data communication in cloud computing network is proposed with multiple layers. The proposed BVDNP-CS method consists of one input layer, three hidden layers and one output layer. The BVDNP-CS method in the first layer obtains the data from personal cloud dataset. In the first hidden layer, with security parameterobtained as input, system parameters and master key are generated by means of Multi-linear mapping. In the second hidden layer, three keys, i.e., partial private key, private key and public key are generated by utilizing Bernstein–Vazirani Key Generation, therefore reducing the overhead incurred.