An Approach of Secret Sharing Technique Based on Convolution Neural Network and DNA Sequence for Data Security in Wireless Communication

Now-a-days the information security is the prime factor in wireless communication. One of the major and fruitful security techniques in cryptography is data encryption with keys. The efficiency of the encryption depends on the resistance ability of the encryption key and the encryption algorithm. Threshold cryptography is one type of cryptographic technique where the information (message and encryption key) is divied into some specific number of shares and on the contrary the information can only be reconstructed by accumulating an allowable set of shares. In this paper we have presented convolution neural network with DNA sequence based (n, n) threshold cryptographic technique. Convolutional Neural Network is one of the important deep neural networks and it has multiple layers. Here pooling layer is used for key generation purposes. We have used maxpooling operation and average pooling operation for encryption key generation and authentication purpose. A new mask generation algorithm is introduced for ‘n’ numbers of share. This mask generation algorithm is based on unit matrix of order. n x n, AND operation and OR operation between plain text and key. This threshold value is also necessary for reconstraction of information. DNA sequence is used for imposing the nonlinearity in cipher text. Different types of experimental results and its analysis prove that the scheme has great potential and ability to achieve the authenticity and confidentiality in any data security platform.