Study and Analysis of Various Cloud Security, Authentication, and Data Storage Models: A Challenging Overview

In recent days, widespread acceptance of cloud data storage applications increases various privacy problems and security problems. Outsourced data security is considered the main confrontation for cloud clients because of data control loss. This review presents a detailed survey of 50 research papers presenting privacy preservation approaches, namely authentication-based, cloud security-based, data storage-based, data security-based, and encryption-based techniques. The analysis is considered based on the categorization of approaches, dataset employed, utilized software tools, published year, and the performance metrics are discussed. Furthermore, problems raised in existing privacy preservation techniques are elucidated in the research gaps and problems section. The future work of this study is based on the research gaps and problems recognized from present research schemes. Additionally, JAVA software language is widely utilized for implementing privacy preservation models, and the Amazon access sample database is a commonly employed dataset for the privacy preservation approach.


INTRODUCTION
Cloud computing is an important task for sharing resources, which includes infrastructures, business processes, applications, and so on.Cloud computing is a border formation of communication convergence.Generally, cloud computing encounters numerous challenges.One of the main challenges in cloud computing is security, which has to turn into the popularization of cloud computing and restrictive factor.If appropriate security measures are not taken into account to the data transmission and cloud operation, the data is at higher risk than while storing or operating in local repositories.Therefore, the cloud process offers data security and application security if the device is missing or damaged (Varadharajan & Tupakula, 2014).In recent days, cloud computing has been paid great attention in Information and Communication Technologies (ICT).Moreover, cloud computing is a data computation or data off-site to internal or exterior, position transport facility or service provider and its moving services (Torkura, et al., 2020).Security is most significant in cloud systems for several business critical computations.Infrastructure as a Service (IAAS) cloud inherits several security concerns linked to utilized technologies.However, the integration of these technologies and a large number of user-sharing resources generate new security concerns.The key element to permit cloud operators and unlimited scaling potential is resource sharing (Banyal, et al., 2013).In a cloud computing system, security is concerned with two layers in software attacks.
Moreover, cloud storage is a raising scheme in the cloud computing model and it enables the user to accumulate their data and access them any time and any place.If the data is entered in the cloud, then users lost their control over data.Therefore, the necessity of control increases as challenging and complex issues for reliability and data storage privacy in the cloud (Jakóbik, 2020).Along with this, security involves major three dimensions, namely confidentiality, integrity, and availability.Recently, public auditing of cloud data looking at integrity guarantee is a progressing research problem.Attribute-Based Encryption (ABE) is a developing cryptographic technique to provide fine-grained access control with confidentiality and outsourced data (Ismail & Islam, 2020).Various researches are focused to strengthen security protection and privacy preservation in the cloud environment (Megouache et al., 2020).Apart from this, various cryptographic approaches were introduced to address privacy problems and potential problems, like data sharing protocols, data public auditing systems, data possession methods, key management, secure data storage systems, privacy-preserving systems, and so on.In addition, previous studies mainly focused on the authentication process to recognize the legal user.Apart from this, many types of research mainly focused on various partly homomorphic cryptosystems for permitting various insufficient operations (Gill, et al., 2020).The advantages of cloud computing receive a lot of attention from both researchers and practitioners, as well as its adoption amid organizations is on the demand.
The major intention of this survey is to analyze several privacy preservation methods.Based on the categorization of review, existing privacy preservation approaches are separated into authenticationbased technique, cloud security-based method, data security-based scheme, data storage-based method, and encryption-based approach.This survey is developed by considering employed software, utilizes dataset, classification of various privacy preservation methods, and so on.Similarly, the computational time is considered as a performance evaluation for the privacy preservation system.The negative aspects available in existing review papers are clearly elucidated in the research gaps and issues section.Hence, the research gaps section is considered as a motivation for future work for effectual privacy preservation approaches.Moreover, the major objective of this paper is to discuss some crucial research topics associated with cloud security and thus assist developers and researchers to better realize the advantages of the cloud computing field and contribute to its development.
The survey paper is arranged as below: Section 2 describes a review of various cloud security, data storage, and authentication models, and section 3 describes research gaps and issues of present authentication, cloud security, and data storage schemes.Section 4 depicts the analysis of authentication, data storage, and cloud security approaches with respect to utilized software, year of publication, toolset employed, performance metrics, and finally, the conclusion of the paper is in section 5.

LITeRATURe SURVey
These research works contain various approaches adopted for privacy preservation are elaborated in this section.The categorization of privacy preservation models is displayed in figure 1.Here, different schemes, like authentication-based approach, cloud security-based approach, data security-based approach, data storage-based technique, and encryption-based approach are developed for the privacy preservation model.The challenges present in these techniques are considered to motivate the researchers for developing a novel privacy preservation process.

Categorization of Privacy Preservation Techniques
The research works considering several approaches used for privacy preservation methods are illustrated in the below sub-sections.

Authentication-Based Approaches
The research that employed authentication-based techniques for privacy preservation is illustrated in this section.Prabha & Saraswathi, (2020) developed multifactor authentication, named Suppressed K-Anonymity Multi-Factor Authentication Based Schmidt-Samoa Cryptography (SKMA-SC) approach for privacy preserved data access in cloud computing.The developed method mainly involves three key processes, like registration, authentication, and data access.Zhou, et al., (2019) devised a lightweight-based authentication method for data access in cloud computing.Here, lightweight crypto modules, like exclusive or operation and one-way hash function was included in this method for better efficiency.This method avoids computation burden and is effective for resource-limited objects.Veerabathiran, et al., (2020) introduced a Homomorphic Proxy Re-Encryption (HPRE) scheme in cloud computing.The basic trial was made in this method for exhibiting lawfully unauthorized access.In this method, cloud users used technical skills for assisting unauthorized users.Wazid, et al., (2020) developed a lightweight authentication approach, termed LAM-CIOT in cloud computing.Here, an authenticated user was used to access the data in a cloud environment.Moreover, the formal security analysis was performed using a Real or Random (ROR) scheme, by the Automated Validation of Internet Security Protocols and Applications (AVISPA) model.A multi-cloud architecture method for ensuring data integrity and authentication was introduced in Megouache, et al., (2020).This method was developed for solving security problems in cloud systems.At first, the private virtual network was employed for data security.At last, this algorithm was realized for recognizing the integrity of data distributed on different clouds of the system.An authentication system based on Paillier using a difference function and a Homomorphic cryptosystem was devised in Falmari & Brindha, (2020).In this method, the homomorphic function was computed using encrypted data and lacking ciphertext decryption.Furthermore, the authentication process was performed in the encrypted environment.Liu, et al., (2014) developed a Shared Authority-based Privacy-preserving Authentication process (SAPA) in cloud computing.Initially, an anonymous access request matching method was applied for obtaining shared access authority.Then, attribute-based access control was implemented for realizing the user and data fields.A multi-factor authentication method for cloud computing was introduced in Banyal, et al., (2013).This developed authentication approach was integrated with the traditional authentication method for obtaining an effective system.This method was verified through the cloud access management method, and it authenticates the user by various factors.Tsai & Lo, (2015) devised a privacy-aware authentication system in cloud computing.This method offers better security and ease for users' access in a cloud computing environment.The security strength of the developed authentication approach was improved using dynamic nonce generation and bilinear pairing cryptosystem.Hao, et al., (2011) developed a time-bound ticket-based mutual authentication protocol for cloud computing.This developed protocol obtained better mutual authentication among the client and server.The relationship between the client smart card and the digital ticket was employed for preventing user masquerade attacks.An enhanced mutual authentication approach for cloud computing was introduced in Jaidhar C D (Jaidhar, 2013).This developed approach verified mutual authenticity among cloud servers and users for producing session keys.This authentication method effectively resists various attacks, such as parallel session attacks, reply attacks, offline password guessing attacks, denial of service attacks, and insider attacks.
Namasudra & Roy, (2017) developed a secure authentication system for a cloud computing environment.This authentication method was developed based on Chebyshev chaotic maps.This scheme considered various security features, like forward security, two-factor security, freedom of password change, mutual authentication, and scalability of login for the authentication process.Belguith, et al., (2020) devised an accountable privacy-preserving attribute-based structure, named Ins-PAbAC.This method integrates attribute-based approaches and attribute-based encryption to secure outsourced data through public cloud servers.An optimized public auditing and data dynamics system for data security in cloud computing was introduced Singh & Pasupuleti, (2016).This method was mainly developed for data integrity verification in the cloud storage server.This system includes three steps, namely setup stage; dynamic data update stage, and third party auditing stage.

Cloud Security-Based Approaches
This section describes the cloud security-based methods for the privacy preservation process.Ismail & Islam, (2020) developed Security Transparency and Audit Framework (STAF) for cloud security.This developed approach involves supporting auditing tools to select cloud service providers and enables security transparency using predefined user needs.
Halabi & Bellaiche, (2018) designed a broker-based structure for controlling cloud security.Initially, the measurable, quantitative, and standard structure was considered for representing agreement.After that, the selection and evaluation method was employed for computing suitable trade-offs among security triad attributes.At last, Security Service Level Agreements (Security-SLA) monitoring method and violation remediation and prediction process were employed.Modic, et al., (2016) introduced the Moving Intervals Process (MIP) for real-time cloud security estimation.Here, various fundamental mathematical principles, like fast Quantitative Hierarchical Process (fQHP) were applied for decreasing computational complexity.After that, a cloud security assessment method was developed for ranking cloud service providers.Lei, et al., (2020) modeled a service recommendation method for enhancing cloud security.Here, a system-based transfer learning scheme was employed for enhancing cold start issues.In addition, extra knowledge was applied from the general domain for improving data security.Almorsy, et al., (2011) developed a collaboration-based technique for cloud security.Here, the collaboration between key cloud stakeholders was improved for sharing needed information.The developed cloud structure mainly includes two behaviors, such as multi-tenancy and elasticity.Varadharajan & Tupakula, (2014) devised security architecture for providing flexible security in a cloud environment.This developed approach provides baseline security to a cloud provider for protecting cloud infrastructure.Consequently, cloud provider offers auditing processes and it guarantees physical security of cloud infrastructure for reducing various hardware-based attacks.Choi & Choi, (2019) developed an ontology-based security context reasoning approach for cloud security.This method contains a data center, multi-service edge, clouds, and other features for the security process.At last, the JAVA platform was exploited to carry out the simulations.A service-based trust management classifier technique for cloud security was introduced Banerjee et al., (2013).Moreover, the constant value was introduced for the classification process, which transmits an access request to the equivalent cluster.Jakóbik, (2020) introduced Stackelberg game-based method for automatic security decisions in cloud computing.In addition, Artificial Neural Network (ANN) was also included for identifying utility functions based on various pipelines.

Data Security-Based Approaches
The data security-based approaches employed for the privacy preservation process are elaborated in this section.Mohamed, et al., (2012) developed a data security method based on cloud computing architecture.Here, the application software was performed for choosing appropriate and high-security encryption methods.The developed software produces evaluation for chosen various encryption approaches.This introduced data security method resolves cloud user security issues, and it assists cloud provider.A combined approach for ensuring data security in cloud systems was introduced in Sood, (2012).This developed structure provides whole security to data in the cloud computing process.Therefore, various methods are employed for protecting significant information from unauthorized parties.Chadwick, et al., (2020) modeled cloud edge-based data security structure in cloud computing.Here, five-level trust methods were employed for cloud edge-based data-sharing infrastructure.In addition, sanitization was executed through an edge device or cloud service provider based on the trust level and the implementation was performed in the JAVA.Chang & Ramachandran, (2015) developed Cloud Computing Adoption Framework (CCAF) for cloud computing.This system developed an integrated solution for checking every data.The Business Process Modelling Notation (BPMN) was employed for simulating data.Lenka & Nayak, (2014) developed Rivest Shamir Adleman (RSA) encryption algorithm in cloud computing.Here, the digital signature approach and RSA technique were combined easily with every cloud computing characteristic, such as Iaas, SaaS, and PaaS.

Data Storage-Based Approaches
This section presents the data storage-based approaches gathered from several privacy preservation research works.Wang, et al., (2010) introduced a privacy-preserving public auditing system for data storage security in cloud computing.Here, random masking and homomorphic authentication were performed for assuring Third Party Auditor (TPA).
An intelligent cryptography procedure was developed by Li, et al., (2017) for data storage in cloud computing.This developed approach separates files and accumulated the data in distributed cloud servers.Here, Efficient Data Conflation (EDcon) technique, Secure Efficient Data Distributions (SED2) approach, and Alternative Data Distribution (AD2) technique were included for the sensitive data retrieval process.Yang & Jia, (2012) modeled proficient and secure dynamic auditing procedures for data storage in cloud computing.The developed method guarantees data privacy based on bilineraity property and cryptography model of bilinear pairing.Wang, (2017) introduced Sata secure Storage scheme-Based on Tornado codes (DSBT) for data security in cloud computing.
Here, removal codes and symmetric encryption were integrated for an effective DSBT system.In this method, a boot password was employed for solving traditional data encryption in key management and preservation.Sermakani & Paulraj, (2020) developed a data auditing and data storage approach for data security in cloud computing.The retrieval approach and secure data storage scheme, named Advanced encryption standard and Triple Data encryption standard-based Secured Storage and Retrieval Algorithm (ATDSRA) was developed to accumulate public and private cloud user's on cloud dataset.In this method, TDES was applied to encrypt and merge the input data.
Wang & Zhang, (2020) modeled Hadoop Distributed File System (HDFS) for optimizing cloud computing data access storage techniques.The developed approach was separated into three segments, namely information access data access, computation, and storage.Universal Generation Function (UGF) system for a distributed data storage system was introduced in Peng et al., (2020).The joint optimization of data part allocation and distributed data storage approach was considered in this method for increasing system dependability.The identity-based data storage approach was developed on cloud computing by Han et al., (2013).The identity-based proxy re-encryption method was applied for shifting the trouble of various file management from owner to proxy server.

Encryption-Based Approaches
This section explained encryption-based schemes in the privacy preservation process.Grover & Kaur, (2016) modeled an encryption-based approach for data security.This approach was divided into two sections, namely, file uploading and file downloading.At first, input files were compressed for improving the encryption process in terms of time and space requirements.After that, symmetric key production process was executed for creating a key for every separate file.The produced key was utilized for encrypting the file in the encryption stage.Sachdev & Bhansali, (2013) introduced an advanced encryption standard approach for cloud computing security.The encryption process was employed through four stages, namely, shift rows, mix columns, sub bytes and add round key.In the sub bytes step, every byte was replaced with another byte using a lookup table.Namasudra et al., (2020) developed Deoxyribonucleic Acid (DNA)-based data encryption approach for improving data security.This approach was introduced based on 1024-bit DNA with the user's public key, data owner's private key, and secret key.Arora et al., (2013) modeled an encryption-based approach for securing user data in cloud computing.Blowfish technique with the single key was utilized for decryption and encryption of messages.The familiar public key approach named RSA was executed for the encryption and decryption process.Sajay et al., (2019) introduced a hybrid encryption technique for data security in the cloud.The major aim of encryption techniques was to store or secure a large amount of information in cloud computing.This method integrates blowfish encryption and homographic encryption technique for improving the security of cloud data.
Namasudra, ( 2019) developed an improved attribute-based encryption scheme for data security in cloud computing.Here, Distributed Hash Table (DHT) network, Attribute-Based Encryption (ABE), and Identity-based Timed Release Encryption (IDTRE) were included for data security.Initially, resources or data were encrypted through encrypted data and attributes of users to separate extracted ciphertext and encapsulated ciphertext.Then, the IDTRE technique was utilized for encrypting decrypted keys and integrates ciphertext of the key with the extracted ciphertext to produce ciphertext shares.Shahid et al., (2020) designed Proficient Security over Distributed Storage (PSDS) approach for data security in the cloud.Initially, the input data were obtained from the data owner, and input data were divided into two groups, such as sensitive and normal.Here, every section was encrypted and distributed, while normal data was uploaded on the single encrypted structure.

Other Privacy Preservation Approaches
In this section, other privacy preservation techniques are depicted below.Pal et al., (2011) developed a trusted and collaborative agent-based two-tier structure to ensure cloud security.Here, the cloud service provider domain and broker domain were implemented as two-tier architecture.Gill et al., (2020) modeled a game-theoretic method for cloud security.The specific selection of detection components directs to a decline in energy consumption, thus it increases the effectiveness of the defender system.Moreover, this approach was divided into two sections, namely non-cooperative games, and cooperative games.The Risk Driven Fault Injection (RDFI) method for cloud security was introduced in Torkura et al., (2020).This method mainly applied chaos engineering principles for cloud security, and it depends on a feedback loop for monitoring, analyzing, executing, and planning security fault injection movement using the knowledge base.Li et al., (2010) designed a domain-based trust method for the security system in a cloud environment.This method separates cloud providers' resource nodes into identical trust domains.This method has five layers, such as security abstract layer, security application layer, physical security layer, system security layer, and trust management layer.GR & Reddy, (2012) developed soft computing approaches for the security system in cloud computing.In this method, the reputation management system was controlled with cloud computing for ensuring the security of data.Khan & Qazi, (2019) designed Elliptic Curve Cryptography (ECC) for data security in cloud computing.Various approaches, such as Euclidean approaches and Euler's phi function was employed for finding the inverse of elements in a particular set.The recent cryptosystem utilizes hybrid encryption where keys were exchanged through asymmetric encryption.
A cryptosystem based on a Genetic Algorithm (GA) for cloud data security introduced in Tahir et al., (2020).The GA was employed to produce keys for the decryption and encryption process.Moreover, the cryptographic scheme was combined for ensuring the integrity and privacy of data.Meanwhile, this method was applied in both images and text.In 2019, Ferrer, et al., (2019) worked on the technologies which permit the storage privacy-aware outsourcing as well as sensitive data pro-cessing to public clouds.Particularly, masking techniques for outsourced data was reviewed on the basis of the data splitting and anonymization, besides to cryptographic techniques.In 2019, Aziz, et al., (2019) worked on the various privacy and security-related issues, which revolves around human genomic data.Additionally, a few cardinal cryptographic concepts were explored that brought efficiency in data computation of secure and private genomic.

ReSeARCH GAPS IDeNTIFIeD
This section illustrates the research gaps and issues faced by existing privacy preservation approaches.This study is performed systematically by considering the scientific basis in the field of privacy preservation approaches.The research problems faced by authentication-based approaches are explained below.The SKMA-SC method was developed in (Prabha & Saraswathi, 2020) for privacypreserving in cloud computing.However, this method was not included cryptography approaches for improving confidentiality and data integrity in the cloud.The lightweight-based authentication scheme was developed in (Zhou et al., 2019) for cloud computing, although this method failed to improve the computational cost.In (Veerabathiran et al., 2020), HPRE protocol was developed for cloud computing, but still, this system has not enhanced the accuracy of input data.The LAM-CIOT technique was utilized in (Wazid et al., 2020) for cloud computing, even though this method failed to decrease the computational cost.In (Megouache et al., 2020), a multi-cloud architecture approach was devised for ensuring data integrity and authentication.However, this method was not included multi-agent systems for improving efficiency.The secure authentication system was introduced in (Namasudra & Roy 2017) in a cloud computing environment.Even though, this method failed to evaluate execution efficiency based on the dissimilar size of data.In (Belguith et al., 2020), Ins-PAbAC approach was developed for data security in cloud computing, but this method was not employed decryption outsourcing methods for achieving better computation cost on the data user side.
The gaps and problems recognized by cloud security-based methods for privacy preservation are discussed below.In (Ismail & Islam, 2020), STAF was designed for cloud security systems, but still, this method has not enhanced the reliability by automation process.The broker-based structure was designed in (Halabi & Bellaiche 2018) for cloud security, but still, this method was not utilized a real database for evaluating cloud service provider security.In (Modic et al., 2016), MIP was developed for cloud security however, this method was failed to combine assessment processes for an effective outcome.The service recommendation technique was developed in (Lei et al., 2020) for cloud computing, but the method failed to utilize a huge database for enhancing the performance of the system.The collaboration-based approach was developed for cloud security in (Almorsy et al., 2011) but, this method was not derived configuration parameter values for a better security system.The ontology-based security context reasoning system (Choi & Choi, 2019) was failed to define interference rules for several contexts.In (Banerjee et al., 2013), a service-based trust management approach was employed for cloud security, however, this method failed to consider various sets of values for easy computation.
The research problems faced in data security-based techniques are illustrated below: The cloud edge-based data security approach was developed in (Chadwick et al., 2020) for data sharing, but still, the execution process is complex and it needs security experts to start policy templates.In (Chang & Ramachandran, 2015), CCAF was developed for data security with cloud computing, but the method failed to improve the performance and does not reduce the execution time.The RSA approach (Lenka & Nayak, 2014) was developed for improving data security in cloud computing.However, this method was not improved encryption techniques for more data security in cloud computing.The research gaps and issues identified by data storage-based approaches for privacy preservation are considered below: The privacy-preserving public auditing system (Wang et al., 2010) was introduced for data storage security in cloud computing.However, this method was not performed multiple auditing tasks for better efficiency.The intelligent cryptography process was introduced in (Li et al., 2017) for the data storage process in the cloud, although this method was not included secure data duplication for increasing data availability level.In (Wang, 2017), DSBT was introduced for data storage in cloud computing, although this method lost the fragmentation, which directs to the incompletion of a huge amount of files.The data auditing and data storage approach was introduced in (Sermakani & Paulraj, 2020) for data security in the merged cloud.However, this method was not included an efficient cryptographic scheme to offer more security in cloud data accesses and cloud data.The HDFS was developed in (Wang & Zhang, 2020) to optimize cloud computing data access storage approaches.However, this method does not consider the data access storage optimization technique for better accuracy.The method in (Peng et al., 2020) failed to apply data partitioning techniques for effective data security.
The research gaps identified through encryption-based techniques for privacy preservation are explained below: The encryption-based approach for data security was introduced in (Grover & Kaur, 2016), but still, this method does not include a parallel encryption process for improving performance.The DNA-based data encryption scheme (Namasudra, et al., 2020) was developed for data security, but this method was failed to enhance the authentication process in cloud computing.The encryptionbased approach was executed in (Arora, et al., 2013) for data security in cloud computing.However, the developed RSA approach consumes high execution time and huge memory space.In (Sajay et al., 2019), a hybrid encryption technique was developed for data security in the cloud, but this method failed to solve cloud storage problems.In (Namasudra, 2019), an improved attribute-based encryption technique was introduced for data security, but still, this method was not employed other intelligent techniques for effective data communication.The PSDS approach was employed in (Shahid, et al., 2020) for data security in the cloud, but this method was failed to reduce computational time.
The research problems faced by other privacy preservation techniques are discussed below: In (Pal, et al., 2011), a trusted and collaborative agent-based two-tier method was developed for cloud security, although this method failed to include malicious activity identifying techniques for preventing unauthorized access to cloud data.The game-theoretic approach was developed in (Gill, et al., 2020) for cloud security, but this method was not executed in a real-time system.In (Torkura, et al., 2020), the RDFI method was developed for cloud security, but this method failed to execute an intelligent recovery scheme to decrease time overhead.The domain-based trust method was designed in (Li, et al., 2010) for cloud security systems.Even though, this method was not identified potential security risks.The ECC approach was developed for data security in the cloud, but this method was not concentrated on storage management systems.In (Tahir, et al., 2020), cryptosystem-based GA was employed for cloud security, although this method was not developed two-way crossover for encrypting other types of data, like video, audio and etc, and also failed to reduce space complexity problem.

ANALySIS AND DISCUSSION
This section explains the analysis and discussion of several privacy preservation models using various research papers in terms of categorization of approaches, dataset utilized, publication year, software employed, and performance metrics.

Analysis Based on Approaches
This section illustrates the analysis using numerous privacy preservation methods.The approaches developed for the privacy preservation process are shown in figure 2. From figure 2, it is noted that 28% of the research papers used authentication-based approaches and 18% of the research were developed cloud security-based approaches.Moreover, data storage-based approaches were employed by 16% of researchers and 10% of the research work was used data security-based approaches, and 14% of researchers employed encryption-based approaches.From the analysis, it noted that authenticationbased schemes are the most commonly employed techniques for privacy preservation.

Analysis Using Publication year
This sub-section depicts the review on the basis of published years in which 50 research papers are analyzed for cloud security, authentication, and data storage models.The analysis using published year is depicted in Table 1.Out of the 50 papers surveyed, more research papers were published in the year 2020.Then, 8 research papers were published in the year 2019.Therefore, this analysis clearly elucidates that recently cloud security became very popular for its use in several fields.

Analysis in Terms of Toolset
This subsection depicts the toolset employed by authentication, cloud security, and data storage methods.Figure 3 displays the analysis in terms of the utilized toolset.The software toolsets used in research papers are MATLAB, Python, JAVA, C language, oracle, NS-2 simulation tool, .Net, and cloud sim simulator tool.From figure 3, it is clearly shown that JAVA software is repeatedly used software tool for privacy preservation schemes.

evaluation in Terms of employed Datasets
This section described the analysis performed with respect to datasets used by distinct research works.Figure 4 displays various databases employed for cloud security, authentication, and data storage methods.The most commonly used privacy preservation models are the Book-Crossing  4, it is well known that the most frequently used dataset is the Amazon access sample dataset.

Analysis in Terms of evaluation Metrics
The analysis made using the performance metrics is discussed in this subsection.The performance metrics considered are accuracy, Root Mean Squared Error (RMSE), Mean Square Error (MSE), Recall, Precision, F-measure, computational complexity, communication cost, computational time, throughput, decryption time, encryption time, and response time.Based on Table 2, it is computed that the computational time and communication cost are frequently preferred performance metrics.

Analysis Based on the Values of Performance Metrics
The analysis based on performance metrics value is elaborated in this section.The analysis based on computational time is described in the subsection.

CONCLUSION
In this study, a systematic way analysis on various privacy preservation approaches was presented.This research was specially trend in the cloud security field.Therefore, this work analysis other studies papers, which were collected from 50 research works and collected papers were categorized in terms of various methods, such as authentication-based approaches, data storage-based approaches, data security-based approaches, cloud security-based approaches, and encryption-based schemes.In this study, the gaps and issues faced by the existing research papers were discussed.Also, this survey recommends the major future scope for privacy preservation approaches by considering several research gaps.In addition, the analysis of the survey was elaborated with respect to categorization techniques, software toolset employed, used datasets, and evaluation metrics.From the analysis, it is

Figure 1 .
Figure 1.Categorization of privacy preservation model

Figure 2 .
Figure 2. Analysis based on approaches

Figure 4 .
Figure 4. Analysis based on employed datasets

Figure 3. Analysis using employed software dataset
, NSL-KDD cup dataset, the National vulnerabilities database, SPECjvm 2008, Unified host and network dataset, Amazon access sample dataset, Amazon EC2, CityPulse dataset, and Cassandra database.From figure