Ensuring confidentiality and privacy of cloud data using a non-deterministic cryptographic scheme
2. Literature review
Several research has proposed varying cryptographic schemes aimed at ensuring cloud data confidentiality and privacy. Amongst them is the work suggested by Huang et al. [14]. In their work, they proposed an i-OBJECT scheme to ensure data confidentiality that depended on the fragmentation, decomposition of information, and the spread of divided data to distorted cloud storage units. Their approach however did not thoroughly investigate the security against different configurations in the cloud which gave their system high security but with increased execution time. On the other hand, A Hard Decisional Composite Residuosity Assumption scheme which is an enhanced function of the Pailler encryption algorithm was proposed by El Makkaoui et al. [19] to ensure the confidentiality of data on the cloud. Their proposed algorithm’s execution time was high. Jain and Kumar proposed a homomorphic cryptographic scheme to boost customers’ conviction regarding the confidentiality of data. Their system allowed for data updates even in the encrypted form without the need for a security key from the cloud service provider. Their system resulted in a high execution time [20].
The work of Zhang et al. [21] proposed the use of a cryptographic scheme using a pairing-based algorithm based on blockchain that generates records that can resist tampering with records of patients to attain data privacy. Their system allowed all auditors on the system to verify the validity of the records but their contents were encapsulated. On the other hand, their approach failed to consider the security of e-health records under a cloud-assisted project and also depicted a high execution time in the data processing stages. Zhang et al. [15] again proposed an attribute-based access control scheme that is decentralized to achieve data confidentiality on the cloud. Their scheme helped to ensure repudiation which allowed for the generation of a secret key without an idea from the users of the system. However, due to the non-uniqueness of the attribute key, unauthorized users can decipher plaintext which increased the execution time as a result of complicity, which has a serious effect on the security of data.
Huang et al in achieving the same objective as Zhang et al. [22] proposed the use of a Lagrange interpolation-based control system to achieve data confidentiality of patient records on the cloud. Their system achieved this through the use of an authority-based scheme to access health records which increased the struggle in breaking the security of the database and accessing et health information. However, their approach had compatibility of systems and management of access problems as a result employed a lot of iteration which increased its execution time.
Rizwan et al. [23] proposed the use of Modular Encryption Standards (MES) integrated with the augmentation of condition-centric risk monitoring aiming to achieve confidentiality of health records. The confidentiality of data was attained by providing layered architecture of the health records. Making any decision regarding risk strategies of the MES is aided by a machine learning algorithm grounded using a Fuzzy Inference System integrated with Neural Networks. This system provides security against insider and outsider attacks by providing five variant keys for encryption. Their system however was not tested on other data types like image, audio, and video. Again there was proportionality between the data size and the execution time when textual data files were used.
Jain et al. [24] proposed the use of Secured Map Reduce to ensure the privacy of data on the cloud by introducing a layered interface between Hadoop Distributed File System as well as Map-Reduce Layer. Their architecture provided privacy, solved expansion concerns in privacy, and ensured data mining tradeoff based on privacy utility but the iteration of the processes influenced the execution time negatively.
Al‐Balasmeh et al. [25] also ensured data privacy and information over vehicular cloud networks (VCNs) through the use of the data and location privacy (DLP) framework which secured the anonymity of personal data by providing location aided by obfuscation technique. In their work, much concentration was not given to securing loaded geo-fence storage infrastructure because it required many iterations to execute the process which has a negative influence on execution time.
Shivashankar and Mary ensured data privacy and reliability through the use of an enhanced Rider Optimization Algorithm (ROA) called Randomized Rider Optimization Algorithm (RROA). This framework used data sanitization and data restoration. The sanitization of data encapsulates the data from unauthorized users while data restoration is meant for data recovery. As a result of the number of iterations involved in data sanitization and restoration, execution time became proportional to the size of the data [26].
Hasan and Agrawal also proposed a new algorithm to ensure data privacy and confidentiality which was based on a probabilistic cryptographic scheme. Their approach used a single key which made it symmetric. Multiple encrypted data was able to represent plaintext which made it unrealistic in associating Ciphertext with plaintext. Despite the security strength of their proposed algorithm, it was still exponential. Their relation between data size and execution time was also proportional [27].
Gajmal and Udayakumar proposed a blockchain-based algorithm to ensure privacy as well as the utility of health data. The data privacy was achieved through the application of Tracy- Singh product aided by Conditional Autoregressive numbers at risk (CAViar)—based Bird Swarm scheme. This was used as an integration of BSA and CAViar to generate the privacy-preserving units. Their algorithm was effective but not efficient because the privacy percentage indicated a linear relationship between data size and time complexity [28].
The work of Shen et al. [29] proposed a proxy re-encryption scheme and oblivious random access memory that support the sharing of data on the cloud for data sharing. The encrypted data resulting from the implementation of the proposed algorithm allows the group to access and save the data resulting in the security of data sharing. Xu et al. [30] proposed a certificateless auditing algorithm aimed at securing of sharing data and privacy of medical records. The execution times of the proposed algorithm are lower as compared with other state-of-the-art algorithms but had higher security and are also appropriate for sharing data on clouds.
In summary, the methodologies reviewed ensured confidentiality and privacy of data on the cloud. Despite ensuring data privacy and confidentiality in the cloud to maximize the benefits associated with the use of the cloud, existing schemes do not provide the resilience necessary against hackers. Their execution times were higher and were also proportional to the sizes of the data used in the execution process. This, therefore, requires a more robust scheme to secure data against intruders with low and unpredicted execution times. This present study integrates Good Prime Numbers, Linear Congruential Generator, Fixed Sliding Door Window algorithm, and XOR circuit gate to frame a Non-Deterministic Cryptographic Scheme (NCS) to attain cloud data privacy and confidentiality with low and unpredicted execution time.
Conclusion: So above is the Ensuring confidentiality and privacy of cloud data using a non-deterministic cryptographic scheme article. Hopefully with this article you can help you in life, always follow and read our good articles on the website: W Tài Liệu
