Improving Cloud Data Encryption Using Customized Genetic Algorithm

Abstract

Data Encryption is widely utilized for ensuring data privacy, integrity, and confidentiality. Nowadays, a large volume of data is uploaded to the cloud, which increases its vulnerability and adds to security breaches. These security breaches include circumstances where sensitive information is being exposed to third parties or any access to sensitive information by unauthorized personnel. The objective of this research is to propose a method for improving encryption by customizing the genetic algorithm (GA) with added steps of encryption. These added steps of encryption include the data being processed with local information (chromosome's value calculated with computer-generated random bits without human intervention). The improvement in the randomness of the key generated is based on altering the population size, number of generations, and mutation rate. The first step of encrypting is to convert sample data into binary form. Once the encryption process is complete, this binary result is converted back to get the encrypted data or cipher-text. Foremost, the GA operators (population size, number of generations, and mutation rate) are changed to determine the optimal values of each operator to bring forth a random key in the minimum possible time, then local intelligence is headed in the algorithm to further improve the outcomes. Local Intelligence consists of local information and a random bit generated in each iteration. Local Information is the current value of a parent in each iteration at the gene level. Both local information and random bit are then applied in a mathematical pattern to generate a randomized key. The local intelligence-based algorithm can operate better in terms of time with the same degree of randomness that is generated with the conventional GA technique. The result showed that the proposed method is at least 80% more efficient in terms of time while generating the secret key with the same randomness level as generated by a conventional GA. Therefore, when large data are intended to be encrypted, then using local intelligence can demonstrate to be better utilized time.

Artificial intelligence-based Karawo motif formation using genetic algorithm

“…By defining and implementing a robust fitness function, the algorithm can systematically assess and enhance the quality of generated motifs. The fitness function used is as (1) [25]:…”

“…For the fitness function, it is determined that the search for the optimum value uses the maximum function. Therefore, the fitness function can be rewritten as (2) [25]:…”

A Novel Scheme for Malicious Nodes Detection in Cloud Markets Based on Fuzzy Logic Technique

“…Homomorphic encryption algorithm was used to support the operations in the encrypted domain. GA was customized in [19] for improving data encryption. The generated key is based on altering the population size, number of generations, and mutation rate.…”

Modified ECC for Secure Data Transfer in Multi-Tenant Cloud Computing

“…Each layer of cloud computing is found to highly responsible for specific services in which SaaS layer enable accession to desired software thereby avoids downloading to our system, IaaS controls networks, virtual machines etc., and PaaS offers facility to employ various applications or services with reduced cost and software governing difficulties [5,6]. The users of cloud computing must encrypt the data before sharing their data to the cloud in order to provide proper security from leaking the data [7,8]. The process of controlling the access is the first and vital step in providing the security for preventing the unauthorized accessing of the data that have been shared.…”