Figures
Abstract
The amount of data generated by electronic systems through e-commerce, social networks, and data computation has risen. However, the security of data has always been a challenge. The problem is not with the quantity of data but how to secure the data by ensuring its confidentiality and privacy. Though there are several research on cloud data security, this study proposes a security scheme with the lowest execution time. The approach employs a non-linear time complexity to achieve data confidentiality and privacy. A symmetric algorithm dubbed the Non-Deterministic Cryptographic Scheme (NCS) is proposed to address the increased execution time of existing cryptographic schemes. NCS has linear time complexity with a low and unpredicted trend of execution times. It achieves confidentiality and privacy of data on the cloud by converting the plaintext into Ciphertext with a small number of iterations thereby decreasing the execution time but with high security. The algorithm is based on Good Prime Numbers, Linear Congruential Generator (LGC), Sliding Window Algorithm (SWA), and XOR gate. For the implementation in C#, thirty different execution times were performed and their average was taken. A comparative analysis of the NCS was performed against AES, DES, and RSA algorithms based on key sizes of 128kb, 256kb, and 512kb using the dataset from Kaggle. The results showed the proposed NCS execution times were lower in comparison to AES, which had better execution time than DES with RSA having the longest. Contrary, to existing knowledge that execution time is relative to data size, the results obtained from the experiment indicated otherwise for the proposed NCS algorithm. With data sizes of 128kb, 256kb, and 512kb, the execution times in milliseconds were 38, 711, and 378 respectively. This validates the NCS as a Non-Deterministic Cryptographic Algorithm. The study findings hence are in support of the argument that data size does not determine the execution time of a cryptographic algorithm but rather the size of the security key.
Citation: Dawson JK, Twum F, Hayfron Acquah JB, Missah YM (2023) Ensuring confidentiality and privacy of cloud data using a non-deterministic cryptographic scheme. PLoS ONE 18(2): e0274628. https://doi.org/10.1371/journal.pone.0274628
Editor: Pandi Vijayakumar, University College of Engineering Tindivanam, INDIA
Received: April 7, 2022; Accepted: September 1, 2022; Published: February 9, 2023
Copyright: © 2023 Dawson et al. This is an open access article distributed under the terms of the Creative Commons Attribution License, which permits unrestricted use, distribution, and reproduction in any medium, provided the original author and source are credited.
Data Availability: All relevant data are within the manuscript and its Supporting Information files.
Funding: The author(s) received no specific funding for this work.
Competing interests: The authors have declared that no competing interests exist.
1. Introduction
The increased activities of humans have made communication complex. This has increased the need to secure these activities by securing the communication channels [1]. National Institute of Standards and Technology defines cloud computing as a ubiquitous enabling model based on pay-as-you-go services that allow the sharing of pooled resources and is considered one of the fastest-growing technology in the world [2]. This has become possible because of its associated advantages of larger geographical treatment, low capital expenditure, scalability, and the ability to access its services anywhere using the internet [3]. Outsourcing of data to cloud service providers makes them prone to attacks due to the unsafe channel for the transfer of data from the cloud client to the cloud service provider’s server [4]. Data confidentiality and privacy play a major role in data security. In [5], a model was proposed to provide a suitable scheme that can help reduce the security challenges of privacy and confidentiality in the cloud as shown in Fig 1.
1.1 Cloud service models
Four service models are considered under cloud computing. Software-as-a-Service (SaaS) is a cloud-based computerized innovation that offers ubiquitous access to a web-based service over the Internet on a pay-as-you-use basis to cloud clients [6]. Software is deployed to be used by all entities who are clients of the service provider [7]. Platform-as-a-Service (PaaS), makes available platforms that allow the development of the applications as well as maintenance of the applications [8]. The cloud clients can create, plan, improve and assess the developed applications directly from the cloud and also monitor the development cycle of the applications. Infrastructure-as-a-Service (IaaS) serves as the basis upon which all other cloud services are built. This replaces the traditional data centers in the normal network architecture. The cloud service providers use the IaaS model to provide the infrastructure upon which cloud client can store their resources [9]. A Container-as-a-Service (CaaS) is where developers use a package for their entire programming task. The container contains all the coding needs, run timing, and configuring together with the libraries the system needs to execute a host machine [10].
1.2 Data confidentiality
Data confidentiality is the guarding of data from unauthorized persons and ensuring that their content is kept as secured as possible [11]. (See Fig 2). It is considered one of the basic security requirements for protecting data. This is vital in cloud computing due to the advantages of cloud computing such as offsite storage, multiuser schemes, and third-party utility providers [3]. Data confidentiality as a cloud security challenge occurs due to current security challenges resulting from flaws in systems design and its implementation [3] and [12].
1.3 Data privacy
Data Privacy is also considered information privacy, which considers the appropriate storage or handling of vital or sensitive information. Data privacy is the result of achieving confidentiality. The approach to ensuring data privacy in the cloud can be attained through the use of data security mechanisms shown in Fig 3 [13].
1.4 Problem statement
Researchers over the years have proposed several methods and techniques to address data confidentiality and privacy problems resulting from high execution times and predicted execution times based on data size [14–16]. However, the problem of confidentiality and privacy of data on the cloud with high and projected execution time persists [5,17,18]. Therefore, this paper proposes a new cloud data security scheme that integrates Good Prime numbers, Linear Congruential Generator (LCG), Sliding Window algorithm, and XOR gate. The approach achieves high security with few iterations and also enhances the execution times and makes it unpredictable.
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.
3. Methodology
A Non-Deterministic Cryptographic Scheme (NCS) is proposed to ensure cloud data privacy and confidentiality. The NCS is an integration of Good Prime, Linear Congruential Generator (LCG) [31], Fixed Sliding Window Algorithm (SWA), and XOR logic gate. The scheme is made up of three stages, including key generation, encryption, and decryption. The flow diagram demonstrating the proposed algorithm with the specified three stages is shown in Fig 4.
In this algorithm, two good prime numbers are selected at random. The product of the two good prime numbers is used as the seed value for the LCG. The other variables for the computation of the LCG are obtained based on Eqs 6, 7, 8 and 9 in this methodology. Twelve numbers are randomly selected from the numbers generated from the LCG. A fixed sliding window with sub-array three is applied to the twelve numbers selected. The plaintext is encrypted by computing sj (maximum value) [seen in Eq 7], modulus of the sub-array (the ASCII values of the alphabets). The decryption process is computed by calculating the modulus of sj (maximum values) and the sub-array
.
3.1 Key generation
3.1.1 Good prime.
These are prime numbers whose squares are bigger than the product of two prime numbers in the sequence of primes at the same position before and after them [32]. The good prime numbers are generated using Eq 1.
(1)
where n is the list of prime numbers, P(n−i) is the preceding prime number from the selected prime and P(n+i) is the subsequent prime number, such that 1≤i≤n−1. As an example, the list of the first five prime numbers is; 2, 3, 5, 7, and 11. The first good prime from the first prime numbers will be computed using Eq 1 as; 52>3*7,52>2*11. From this, 5 can be considered a good prime. Therefore the first eight good prime numbers are [5, 11, 17, 29, 37, 41, 53, and 59]. Based on Eq (1), we select two random numbers P and Q, such that P*Q = H, P≠Q, P∈Zk, and Q∈Zk. The resultant H serves as the seed value for the Linear Congruential Generator which serves as the next stage in the key generation process.
3.1.2 Apply linear congruential generator.
Any mathematical formula that results in the generation of categorization of randomized numbers computed based on a sporadic equation is considered Linear Congruential Generator (LCG) [31,33],. In generating the sequence of the values between X1, X2… and 0, m−1, the recursive relation of the numbers is shown by Eq (2).
(2)
Eq 2, work on the conditions such that; m>0, a<m, c<m, Xo<m
where a = multiplier, c = increment, m = modulus, and Xi = seed value. Using the product of P and Q as the seed value for Eq (2) generates a hundred thousand random numbers (100,000). The twelve numbers selected are the computational values for the Fixed Sliding Window Algorithm to generate the maximum and minimum values.
3.1.3 Apply fixed sliding window algorithm.
The sliding window algorithm is used when identifying the results for a range of numbers in an array. This has the objective of converting a nested group of loops into a single loop which helps to reduce the complexity time from O(n2) to O(n). The theory behind the sliding window algorithm is to generate the maximum or minimum units by computing the results continually for a range based on an array given [34].
Apply the Fixed Sliding Window algorithm using Eqs 3, 4, 5 and 6 to generate four (ay, ay1, ay2, ay3) numbers using a sub-array of 3 from the twelve number array depicted in Tables 1 and 2.
Use Eq 7 to select the maximum and minimum numbers from the four numbers computed. Where sj is the maximum value after applying the Fixed Sliding Window Algorithm on the 12 arrays generated?
3.2 Encryption
The XOR gate (⨁) shown in Fig 5, and Eq 8 is applied in the encryption process. In ⨁ gate, there is a combination of gates which results in the complex logic gate that is used more in constructing logic gates for arithmetic circuits, comparators for logic computation as well as detection of errors. In ⨁ gate, a “HIGH” voltage is obtained when all the input terminals are “DIFFERENT” and a “LOW” voltage when the input is all “HIGH” [35].
The ⨁ gate can be represented in Eq 8 as
(8)
Encryption is the conversion of plaintext to Ciphertext and this is achieved by applying Eq 9 to the plaintext.
The plaintext is encrypted by computing sj (maximum value) seen in Eq 7 modulus of the sub-array (). We then computed the results ⨁ xi (the ASCII values of the alphabets) to obtain an eight-bit string for each alphabet.
3.3 Decryption
Decryption is the conversion of Ciphertext to plaintext and is computed based on the formula in Eq 10.
The decryption process is computed by calculating the modulus of sj (maximum values) (see Eq 7) and the sub-array (). Compute the output ⨁ yi to get an eight-bit string and the corresponding ASCII value is obtained to obtain the plaintext.
The proposed algorithm for the proposed scheme as indicated in Fig 4 is shown in Fig 6.
The architectural framework of the proposed algorithm is depicted in Fig 7. The plaintext for the cloud storage is converted to Ciphertext by applying Eq 9. The Ciphertext is then sent through the Internet Service provider’s network for cloud storage. On request for the Ciphertext from the cloud service provider, Eq 10 is applied on the Ciphertext to convert it to plaintext for onward forwarding to the recipient.
4 Implementation, results, and discussion
4.1 Implementation
This section presents the implementation of the non-deterministic theory of the proposed NCS algorithm. The Sliding Window Algorithm (SWA) is applied to the twelve arrays of numbers generated from the application of LCG as shown in Fig 8. The maximum value selected which cannot be predicted under any circumstances gives the NCS algorithm the non-deterministic feature as shown in Fig 8. The message is then encoded by computing sj (maximum value) seen in Eq 7 modulus of the sub-array (). We then computed the results ⨁ xi (the ASCII values of the alphabets) to obtain an eight-bit string for each alphabet as shown in Fig 9.
The deciphering process is calculated by finding the modulus of sj (maximum values) (see Eq 7) and the sub-array (). The output is then ⨁ yi (Ciphertext) to get an eight-bit string and the corresponding ASCII value is obtained to achieve the plaintext as shown in Fig 10.
Data and environment for the experiment.
The experiment was conducted on an i7 Lenovo computer, a 2.10GHz CPU, and implemented using the C# language. Personal data was generated for the experimental work in Figs 11, 12 and 13. A predesigned dataset [36] was used to test the average execution times for NCS. The dataset was again used for the comparative analysis of the execution times for NCS, DES, AES, and RSA based on the data size used in the work of Ali et al. [37]. Personal data size of 4bk was executed using NCS and the corresponding Ciphertext and plaintext are shown in Figs 12 and 13.
4.2 Results
The scheme was tested using a message size of 2n. kb (n∈1, 3, 5, 7, 9). The scheme was run 30 times to generate 30 different encryption and decryption times and their average was computed as shown in Tables 3 and 4 and Figs 14 and 15. The execution times for the comparative analysis for NCS, AES, DES and RSA are shown in Tables 5 and 6 and Figs 14 and 15.
4.3 Discussion
Time complexity refers to the time it takes for an algorithm to encipher a plaintext into a Ciphertext and decipher a Ciphertext into a plaintext. When the scrambling time is faster, this implies the data processing time is faster and such algorithms are considered to be efficient [38]. There have been a series of research to propose algorithms with the lowest execution time with a nonlinear time complexity O(n) [39]. In this paper, a Non-Deterministic Cryptographic Scheme is proposed which is a symmetric algorithm. The algorithm was run 30 times to generate 30 different encryption and decryption times, and their averages were computed to consider the performance metrics of execution times of the algorithm using different data sizes. A comparative analysis was conducted based on the execution times of existing cryptographic algorithms such as RSA [37], AES [37], and, DES [37] as shown in Tables 5 and 6 and Figs 16 and 17.
From Tables 3 and 4, and Figs 12 and 13, it can be deduced that the average encryption time is high for a smaller data size of 2Kilobyte (KB) with an increased decryption time but reduced when the data size was increased to 23KB with an increased decryption average time of 540.0667 Millisecond. There was also an observed reduction in average encryption time using a data size of 25 KB with an increased average decryption time for the proposed algorithm. There was a tremendous reduction in average decryption time with an increased encryption time for the proposed algorithm when the data size was raised to 27 KB. This is supported by the work of [40] that the execution time of an algorithm depends on the function of the size of the key used as the security key. The output of the averages of the proposed algorithm overrides the idea of the works of [37,41,42] indicating that file size is proportional to the encryption and decryption time of an algorithm.
Tables 5 and 6 and Figs 16 and 17 show the comparison between the encryption and decryption time of AES, DES, RSA, and the proposed algorithm NCS. From Table 5 and Fig 16, it can be observed that using a key size of 128 KB, NCS has the lowest encryption time followed by AES, DES, and RSA. This indicates the superiority of AES over other conventional algorithms but not against the proposed NCS algorithm [43]. Again using a key length of 512KB, the proposed algorithm has the lowest encryption time as opposed to RSA which has the highest. This confirms the theory that symmetric algorithms have the lowest encryption time during text file processing [38]. On the other hand, using a key size of 256KB, the decryption time for DES and RSA were the same while the proposed algorithm had the lowest. Contrarily, with a data size of 128kb, the execution times were lower but increased with a data size of 256kb and decreased again when the data size was increased to 512kb for the proposed NCS algorithm. It was justified that data size does not determine the execution time of an algorithm but is dependent on the function of the size of the security key [40].
4.4 Novelty
In an attempt by researchers to propose algorithms with low execution times, their time complexity remains linear [44]. Our proposed algorithm produces a nonlinear time complexity as depicted in Tables 5 and 6 and Figs 16 and 17. The execution times are dependent on the size of the security key and not the data size. For the proposed NCS algorithm, the encryption and decryption times are dependent on the value of sj as shown in Eq 7. This, therefore, influenced the execution times when data size was increased from 128kb, 256kb to 512kb with corresponding encryption times of 129ms, 552ms, and 358ms and decryption times of 38ms, 711ms, and 378ms. This makes our proposed NCS algorithm execution time nondeterministic.
5 Conclusion
The proposed algorithm was run 30 times using a dataset from Kaggle considering the data sizes used in the works of Ali et al. [37] and their averages computed to give execution times performance for the proposed NCS algorithm. The performance evaluation of three conventional algorithms (AES, DES, and RSA) together with the proposed algorithm was also evaluated based on their encryption and decryption times. There were differences between the encryption times for AES, DES, RSA, and NCS. The proposed NCS algorithm had the lowest encryption time with RSA having the highest. With decryption, with a data size of 256KB, the decryption time was higher compared with a data size of 512KB for the proposed NCS algorithm. From the results obtained, the security strength of the proposed algorithm (NCS) is stronger compared with industry-embraced algorithms like AES, DES, and RSA. Future works should be conducted to evaluate the memory consumption rate, CPU usage, and throughput against different file sizes.
References
- 1. P S. M. and Dr.D V., “A Study of Data Storage Security Issues in Cloud Computing,” Bonfring International Journal of Software Engineering and Soft Computing, vol. 9, no. 2, pp. 05–07, Apr. 2019,
- 2. Kumar P. R., Raj P. H., and Jelciana P., “Exploring Data Security Issues and Solutions in Cloud Computing,” Procedia Computer Science, vol. 125, pp. 691–697, 2018,
- 3. Rath A., Spasic B., Boucart N., and Thiran P., “Security Pattern for Cloud SaaS: From System and Data Security to Privacy Case Study in AWS and Azure,” Computers, vol. 8, no. 2, p. 34, May 2019,
- 4. El-Yahyaoui A. and Ech-Chrif El Kettani M. D., "Data privacy in cloud computing," 2018 4th International Conference on Computer and Technology Applications (ICCTA), 2018, pp. 25–28,
- 5. Sharma Y., Gupta H. and Khatri S. K., "A Security Model for the Enhancement of Data Privacy in Cloud Computing," 2019 Amity International Conference on Artificial Intelligence (AICAI), 2019, pp. 898–902,
- 6. Mero J., Leinonen M., Makkonen H., and Karjaluoto H., “Agile logic for SaaS implementation: Capitalizing on marketing automation software in a start-up,” Journal of Business Research, vol. 145, pp. 583–594, Jun. 2022,
- 7. Liu S., Yue K., Yang H., Liu L., Duan X., and Guo T., "The Research on SaaS Model Based on Cloud Computing," 2018 2nd IEEE Advanced Information Management, Communicates,Electronic and Automation Control Conference (IMCEC), 2018, pp. 1959–1962,
- 8. Saraswat M. and Tripathi R. C., "Cloud Computing: Analysis of Top 5 CSPs in SaaS, PaaS and IaaS Platforms," 2020 9th International Conference System Modeling and Advancement in Research Trends (SMART), 2020, pp. 300–305,
- 9. AbdelGhany S. Y. and Mamdouh Hassan H., "Get as you Pay Model for IaaS Cloud Computing," 2018 International Conference on Smart Communications and Networking (SmartNets), 2018, pp. 1–6,
- 10. Hussein M., Mousa M., and Alqarni M., "A placement architecture for a container as a service (CaaS) in a cloud environment", Journal of Cloud Computing, vol. 8, no. 1, 2019. Available:
- 11. Hambouz A., Shaheen Y., Manna A., Al-Fayoumi M. and Tedmori S., "Achieving Data Integrity and Confidentiality Using Image Steganography and Hashing Techniques," 2019 2nd International Conference on new Trends in Computing Sciences (ICTCS), 2019, pp. 1–6,
- 12. Paje R. E. J., Sison A. M., and Medina R. P., “Multidimensional key RC6 algorithm,” Proceedings of the 3rd International Conference on Cryptography, Security and Privacy—ICCSP ‘19, 2019,
- 13. Godi S. and Kurra R., “Novel security issues and mitigation measures in cloud computing: an Indian perspective,” International Journal of Computer Applications in Technology, vol. 58, no. 4, p. 267, 2018,
- 14. Branco E. C., Monteiro J. M., de C R. R. Silva e, and Machado J. C., “A New Approach to Preserving Data Confidentiality in the Cloud,” Proceedings of the 20th International Database Engineering & Applications Symposium on—IDEAS ‘16, 2016,
- 15. Zhang L., Gao X., Kang L., Liang P., and Mu Y., “Distributed Ciphertext-Policy Attribute-Based Encryption With Enhanced Collusion Resilience and Privacy Preservation,” IEEE Systems Journal, vol. 16, no. 1, pp. 735–746, Mar. 2022,
- 16. Prabahar L., Sukumar R., and SureshBabu R., “CCSC—DHKEP: Data Confidentiality Using Improved Security Approaches in Cloud Environment,” Wireless Personal Communications, vol. 122, no. 4, pp. 3633–3647, Sep. 2021,
- 17. Timraz K., Barhoom T. and Fatayer T., "A Confidentiality Scheme for Storing Encrypted Data through Cloud," 2019 IEEE 7th Palestinian International Conference on Electrical and Computer Engineering (PICECE), 2019, pp. 1–5,
- 18. Li D., Wu J., Le J., Liao X. and Xiang T., "A Novel Privacy-Preserving Location-Based Services Search Scheme in Outsourced Cloud," in IEEE Transactions on Cloud Computing,
- 19. El Makkaoui K., Ezzati A., and Beni-Hssane A., “Securely Adapt a Paillier Encryption Scheme to Protect the Data Confidentiality in the Cloud Environment,” Proceedings of the International Conference on Big Data and Advanced Wireless Technologies—BDAW ‘16, 2016,
- 20. Jain A. and Kumar R., “Confidentiality Enhanced Security Model for Cloud Environment,” Proceedings of the Second International Conference on Information and Communication Technology for Competitive Strategies—ICTCS ‘16, 2016,
- 21. Zhang G., Yang Z., and Liu W., “Blockchain-based privacy preserving e-health system for healthcare data in cloud,” Computer Networks, p. 108586, Nov. 2021,
- 22. Huang Y.T., Chiang D.L., Chen T. S., Wang S.D., Lai F.P., and Lin Y.D., “Lagrange interpolation-driven access control mechanism: Towards secure and privacy-preserving fusion of personal health records,” Knowledge-Based Systems, vol. 236, p. 107679, Jan. 2022,
- 23. Rizwan M. et al., “Risk monitoring strategy for confidentiality of healthcare information,” Computers and Electrical Engineering, vol. 100, p. 107833, May 2022,
- 24. Jain P., Gyanchandani M., and Khare N., “Enhanced Secured Map Reduce layer for Big Data privacy and security,” Journal of Big Data, vol. 6, no. 1, Mar. 2019,
- 25. Al‐Balasmeh H., Singh M., and Singh R., “Framework of data privacy preservation and location obfuscation in vehicular cloud networks,” Concurrency and Computation: Practice and Experience, vol. 34, no. 5, Oct. 2021,
- 26. Shivashankar M. and Mary S. A., “Privacy preservation of data using modified rider optimization algorithm: Optimal data sanitization and restoration model,” Expert Systems, vol. 38, no. 3, Jan. 2021,
- 27. Hasan Z., Agrawal C. P. and Agrawal M., ""Online Transaction Security Enhancement": An Algorithm Based on Cryptography," 2019 International Conference on Issues and Challenges in Intelligent Computing Techniques (ICICT), 2019, pp. 1–4,
- 28. Gajmal Y. M. and Udayakumar R., “Privacy and Utility-Assisted Data Protection Strategy for Secure Data Sharing and Retrieval in Cloud System,” Information Security Journal: A Global Perspective, pp. 1–15, Jun. 2021,
- 29. Shen J., Yang H., Vijayakumar P. and Kumar N., "A Privacy-Preserving and Untraceable Group Data Sharing Scheme in Cloud Computing," in IEEE Transactions on Dependable and Secure Computing, vol. 19, no. 4, pp. 2198–2210, 1 Jul-Aug. 2022,
- 30. Xu Z., He D., Vijayakumar P., Gupta B. and Shen J., "Certificateless Public Auditing Scheme with Data Privacy and Dynamics in Group User Model of Cloud-Assisted Medical WSNs," in IEEE Journal of Biomedical and Health Informatics, pmid:34788225
- 31. Ankur Divyanjali and Bhardwaj T., "A dissection of pseudorandom number generators," 2015 2nd International Conference on Signal Processing and Integrated Networks (SPIN), 2015, pp. 318–323,
- 32. Patel M., Patel A. M., and Gandhi R. B., “Prime numbers and their analysis,” Journal of Emerging Technologies and Innovative Research, vol. 7, no. 2, pp. 1–5, Mar. 2020, ISSN-2349-5162.
- 33. Herrero-Collantes M. and Garcia-Escartin J. C., “Quantum random number generators,” Reviews of Modern Physics, vol. 89, no. 1, Feb. 2017,
- 34. Li Y., Wei H.-L., Billings Stephen. A., and Sarrigiannis P. G., “Identification of nonlinear time-varying systems using an online sliding-window and common model structure selection (CMSS) approach with applications to EEG,” International Journal of Systems Science, vol. 47, no. 11, pp. 2671–2681, Mar. 2015,
- 35. Bellizia D., Palumbo G., Scotti G. and Trifiletti A., "A Novel Very Low Voltage Topology to implement MCML XOR Gates," 2018 14th Conference on Ph.D. Research in Microelectronics and Electronics (PRIME), 2018, pp. 157–160,
- 36.
“English to French translations,” www.kaggle.com. https://www.kaggle.com/datasets/digvijayyadav/frenchenglish/metadata (accessed May 09, 2022).
- 37. Ali K., Akhtar F., Memon S. A., Shakeel A., Ali A. and Raheem A., "Performance of Cryptographic Algorithms based on Time Complexity," 2020 3rd International Conference on Computing, Mathematics and Engineering Technologies (iCoMET), 2020, pp. 1–5,
- 38. Thangapandiyan M., Anand P. M. R. and Sankaran K. S., "Enhanced Cloud Security Implementation Using Modified ECC Algorithm," 2018 International Conference on Communication and Signal Processing (ICCSP), 2018, pp. 1019–1022,
- 39. Mohamed A. M. and Minn H., "Low-Complexity Algorithm for Radio Astronomy Observation Data Transport in an Integrated NGSO Satellite Communication and Radio Astronomy System," in IEEE Open Journal of the Communications Society, vol. 2, pp. 2342–2356, 2021,
- 40. Masram R., Shahare V., Abraham J., and Moona R., “Analysis and Comparison of Symmetric Key Cryptographic Algorithms Based on Various File Features,” International Journal of Network Security & Its Applications, vol. 6, no. 4, pp. 43–52, Jul. 2014,
- 41. Bentajer A., Hedabou M., Abouelmehdi K., and Elfezazi S., “CS-IBE: A Data Confidentiality System in Public Cloud Storage System,” Procedia Computer Science, vol. 141, pp. 559–564, 2018,
- 42. Priya K. and ArokiaRenjit J., "Data security and confidentiality in public cloud storage by extended QP protocol," 2017 International Conference on Computation of Power, Energy Information and Commuincation (ICCPEIC), 2017, pp. 235–240,
- 43. Latif I. H., “Time Evaluation Of Different Cryptography Algorithms Using Labview,” IOP Conference Series: Materials Science and Engineering, vol. 745, no. 1, p. 012039, Feb. 2020,
- 44. Liu H., Lu X., Duan S., Zhang Y. and Xiang Y., "An Efficient Oblivious Random Data Access Scheme in Cloud Computing," in IEEE Transactions on Cloud Computing,